sentence1
stringlengths 40
15.9k
| sentence2
stringlengths 88
20k
| label
float64 0.5
0.99
|
|---|---|---|
5. The one or more computer-readable storage media of claim 1 , wherein the method further comprises receiving a third query result determined based upon a search conducted utilizing a query result having a confidence value that is most favorable. | 5. The one or more computer-readable storage media of claim 1 , wherein the method further comprises receiving a third query result determined based upon a search conducted utilizing a query result having a confidence value that is most favorable. 6. The one or more computer-readable storage media of claim 5 , wherein the method further comprises displaying an indication of the spelling utilized in determining a respective query result. | 0.921569 |
11. A computerized method for rendering an online collaborative editable document, comprising: receiving, at a client computing device, a user request for displaying a raw online collaborative editable document having one or more graphical elements and one or more textual elements, wherein the raw online collaborative editable document is configured to be rendered in a graphical format in a web browser and the raw document includes at least one of a presentation slide, a spreadsheet and a word-processing document page, and wherein the user request is received via a web-application installed on the client computing device; retrieving, at the client computing device, a composite document having HTML data and CSS data representative of the one or more graphical elements and the one or more textual elements, wherein the CSS data aligns graphical elements with textual elements matching the alignment of the one or more graphical elements and the one or more textual elements in the raw document; and rendering the composite document in a web browser on the client computing device; wherein the composite document includes a combination of a graphical data file and a textual data file, the composite document including the textual data file having additional data to including the graphical data file, wherein the textual data file includes the one or more textual elements in the document and includes HTML data and CSS data representative of content and appearance of only the one or more identified textual elements; and wherein when rendered by the web browser, the textual data file overlays the graphical data file such that alignment, content and appearance of one or more graphical elements and one or more textual elements of the composite document are the same as the alignment, content and appearance of the one or more graphical elements and the one or more textual elements of the raw document. | 11. A computerized method for rendering an online collaborative editable document, comprising: receiving, at a client computing device, a user request for displaying a raw online collaborative editable document having one or more graphical elements and one or more textual elements, wherein the raw online collaborative editable document is configured to be rendered in a graphical format in a web browser and the raw document includes at least one of a presentation slide, a spreadsheet and a word-processing document page, and wherein the user request is received via a web-application installed on the client computing device; retrieving, at the client computing device, a composite document having HTML data and CSS data representative of the one or more graphical elements and the one or more textual elements, wherein the CSS data aligns graphical elements with textual elements matching the alignment of the one or more graphical elements and the one or more textual elements in the raw document; and rendering the composite document in a web browser on the client computing device; wherein the composite document includes a combination of a graphical data file and a textual data file, the composite document including the textual data file having additional data to including the graphical data file, wherein the textual data file includes the one or more textual elements in the document and includes HTML data and CSS data representative of content and appearance of only the one or more identified textual elements; and wherein when rendered by the web browser, the textual data file overlays the graphical data file such that alignment, content and appearance of one or more graphical elements and one or more textual elements of the composite document are the same as the alignment, content and appearance of the one or more graphical elements and the one or more textual elements of the raw document. 15. The computerized method of claim 11 , further comprising sending the raw document from the client computing device to a server for storage. | 0.588731 |
1. A method comprising: identifying, based on past interactions with a user participating in a dialog with a speech dialog system, an adaptation schema which, when applied to a speech recognition model, increases a likelihood the speech recognition model will recognize misrecognized speech from the user relative to an unadapted speech recognition model; determining, via a processor configured to perform speech recognition, that the user has previously repeated speech inputs based on interactions with the user prior to initiating the dialog, to yield a determination; and adapting, via the processor and based on the determination, the speech recognition model using the adaptation schema before an expected repeat speech input, wherein adapting the speech recognition model further comprises modifying an acoustic model, a language model, and a semantic model. | 1. A method comprising: identifying, based on past interactions with a user participating in a dialog with a speech dialog system, an adaptation schema which, when applied to a speech recognition model, increases a likelihood the speech recognition model will recognize misrecognized speech from the user relative to an unadapted speech recognition model; determining, via a processor configured to perform speech recognition, that the user has previously repeated speech inputs based on interactions with the user prior to initiating the dialog, to yield a determination; and adapting, via the processor and based on the determination, the speech recognition model using the adaptation schema before an expected repeat speech input, wherein adapting the speech recognition model further comprises modifying an acoustic model, a language model, and a semantic model. 2. The method of claim 1 , further comprising recognizing the expected repeat speech input from the user based on an adapted speech recognition model. | 0.883308 |
17. The system of claim 16 , wherein the piece of equipment comprises an interface slot and the voice detection unit is plugged into the interface slot. | 17. The system of claim 16 , wherein the piece of equipment comprises an interface slot and the voice detection unit is plugged into the interface slot. 20. The system of claim 17 , wherein the piece of equipment comprises the telephone gateway and the voice detection unit is plugged into the interface slot of the telephone gateway. | 0.975584 |
1. A computer-implemented method for software testing and validation, the computer-implemented method performing a process of building an abstract model of a software application under test and a process of testing at least one region of the software under test, the method comprising the steps of: enabling a user to define at least one specific region of the software application under test by tagging at least one hierarchical element and at least one child region of the specific region by tagging at least one child element of the at least one hierarchical element; enabling a user to tag test objects within the software application under test; building the abstract model using hierarchical elements, child elements and test objects, the hierarchical elements, child elements and test objects indexing at least one object in the at least one region of the software under test; executing a test script to test the at least one object in the at least one region by navigating the abstract model using one or more of the following hierarchical element, a child element, and a test object; altering the software application under test by relocating or redefining the at least one region of the software application under test; and re-executing the test script to test the at least one object in the at least one region by navigating the abstract model using one or more of the following: the hierarchical element, the child element, and the test object, the re-executing performed without regeneration of the abstract model. | 1. A computer-implemented method for software testing and validation, the computer-implemented method performing a process of building an abstract model of a software application under test and a process of testing at least one region of the software under test, the method comprising the steps of: enabling a user to define at least one specific region of the software application under test by tagging at least one hierarchical element and at least one child region of the specific region by tagging at least one child element of the at least one hierarchical element; enabling a user to tag test objects within the software application under test; building the abstract model using hierarchical elements, child elements and test objects, the hierarchical elements, child elements and test objects indexing at least one object in the at least one region of the software under test; executing a test script to test the at least one object in the at least one region by navigating the abstract model using one or more of the following hierarchical element, a child element, and a test object; altering the software application under test by relocating or redefining the at least one region of the software application under test; and re-executing the test script to test the at least one object in the at least one region by navigating the abstract model using one or more of the following: the hierarchical element, the child element, and the test object, the re-executing performed without regeneration of the abstract model. 5. A computer-implemented method of claim 1 , wherein the abstract model is navigated using a plurality of integers, the plurality of integers being assigned to the hierarchical elements and child elements to thereby indicate the location of the test object. | 0.580557 |
14. The computer program product of claim 9 , wherein the computer readable program, when executed on the system, causes the at least one processing device to identify the first validated code snippet by performing machine learning and natural language processing in combination with code analysis techniques to implement a fuzzy matching algorithm for selecting a first candidate code snippet having first internal extracted features that match second internal extracted features from the first library function. | 14. The computer program product of claim 9 , wherein the computer readable program, when executed on the system, causes the at least one processing device to identify the first validated code snippet by performing machine learning and natural language processing in combination with code analysis techniques to implement a fuzzy matching algorithm for selecting a first candidate code snippet having first internal extracted features that match second internal extracted features from the first library function. 15. The computer program product of claim 14 , wherein the computer readable program, when executed on the system, causes the at least one processing device to implement the fuzzy matching algorithm by performing Abstract Syntax Tree (AST) matching to identify the first candidate code snippet as the first validated code snippet which matches the second internal extracted features from the first library function. | 0.815557 |
1. A method comprising: converting, by a voice recognition system, an acoustic signal, representing a request for location information associated with an item having an association with an aisle and a shelf, into a digital signal; monitoring, by the voice recognition system, for a user-definable setting, wherein the user-definable setting comprises a trigger word transmitted to the voice recognition system; determining, by the voice recognition system, one or more words included in the acoustic signal based at least in part on the digital signal and said monitoring; and providing, by the voice recognition system, feedback to the request for location information based at least in part on the determined one or more words, wherein the feedback includes an aisle identifier that identifies the aisle and a shelf identifier that identifies the shelf. | 1. A method comprising: converting, by a voice recognition system, an acoustic signal, representing a request for location information associated with an item having an association with an aisle and a shelf, into a digital signal; monitoring, by the voice recognition system, for a user-definable setting, wherein the user-definable setting comprises a trigger word transmitted to the voice recognition system; determining, by the voice recognition system, one or more words included in the acoustic signal based at least in part on the digital signal and said monitoring; and providing, by the voice recognition system, feedback to the request for location information based at least in part on the determined one or more words, wherein the feedback includes an aisle identifier that identifies the aisle and a shelf identifier that identifies the shelf. 4. The method of claim 1 , wherein said providing feedback to the request for location information comprises providing auditory feedback to the request for location information. | 0.731781 |
7. A corpus processing agent comprising: a processor; a memory coupled to the processor; an audio capture device; and a processing logic for: monitoring a conversation between one or more speakers; identifying the spoken languages of the conversation; identifying one or more topics being discussed within the conversation; in response to identifying the topics being discussed, creating a plurality of metadata tags for each topic of the conversation, wherein the metadata tags include, for each topic of the conversation, one or more of: a description of the speakers for a portion of the conversation, a description of the languages spoken for a portion of the conversation, a summary of the topic of the conversation for a portion of the conversation, a plurality of links to other related topics of the conversation, and a plurality of links to other related topics of a previously analyzed conversation; storing the metadata tags in a link database; determining a spoken emotional pattern of an autonomously selected topic of the conversation; and creating a corpus of the conversation, wherein the corpus includes a text transcription of the conversation, and also includes an identification of the spoken emotional pattern and metadata tags of the conversation. | 7. A corpus processing agent comprising: a processor; a memory coupled to the processor; an audio capture device; and a processing logic for: monitoring a conversation between one or more speakers; identifying the spoken languages of the conversation; identifying one or more topics being discussed within the conversation; in response to identifying the topics being discussed, creating a plurality of metadata tags for each topic of the conversation, wherein the metadata tags include, for each topic of the conversation, one or more of: a description of the speakers for a portion of the conversation, a description of the languages spoken for a portion of the conversation, a summary of the topic of the conversation for a portion of the conversation, a plurality of links to other related topics of the conversation, and a plurality of links to other related topics of a previously analyzed conversation; storing the metadata tags in a link database; determining a spoken emotional pattern of an autonomously selected topic of the conversation; and creating a corpus of the conversation, wherein the corpus includes a text transcription of the conversation, and also includes an identification of the spoken emotional pattern and metadata tags of the conversation. 8. The corpus processing agent of claim 7 , the processing logic further comprising: creating a paraphrase for each topic of the conversation; and in response to determining the paraphrase, modifying the corpus to include the paraphrase. | 0.541525 |
15. A computer readable storage medium having computer executable instructions that when executed by one or more processors cause one or more computers to: receive at least one query, the at least one query having a plurality of terms including at least one camel case term and at least one value type term; process the at least one query to modify the at least one camel case term into its constituent terms; process the at least one query to modify the at least one value type term into a generic term; store the at least one value type term in memory; create at least one modified query the at least one modified query modified to have the constituent terms at a position corresponding to the camel case term and the generic term at a position corresponding to the value type term; identify a plurality of documents in a knowledge base database that at least proximately match the at least one modified query; determine a distance for each term in the modified query in the plurality of documents; remove from the plurality of documents any documents wherein the distance exceeds a predetermined distance; apply a scoring formula to the plurality of documents to obtain a score for each of the plurality of documents; and rank the plurality of documents according to the obtained score. | 15. A computer readable storage medium having computer executable instructions that when executed by one or more processors cause one or more computers to: receive at least one query, the at least one query having a plurality of terms including at least one camel case term and at least one value type term; process the at least one query to modify the at least one camel case term into its constituent terms; process the at least one query to modify the at least one value type term into a generic term; store the at least one value type term in memory; create at least one modified query the at least one modified query modified to have the constituent terms at a position corresponding to the camel case term and the generic term at a position corresponding to the value type term; identify a plurality of documents in a knowledge base database that at least proximately match the at least one modified query; determine a distance for each term in the modified query in the plurality of documents; remove from the plurality of documents any documents wherein the distance exceeds a predetermined distance; apply a scoring formula to the plurality of documents to obtain a score for each of the plurality of documents; and rank the plurality of documents according to the obtained score. 19. The computer readable storage medium of claim 15 , wherein the computer executable instructions executed by the one or more processors cause the one or more computers to further perform the following: determine that the received query exists in a cache of preexisting queries; and output a ranking of documents for the query from the cache when the query exists in the cache. | 0.526428 |
1. A computer based method for executing a command in a vehicle, comprising the steps of: receiving a voice signal; receiving a gesture signal; performing a voice recognition procedure to generate a voice recognition (VR) score representing a likelihood that the voice signal corresponds to a first voice command; performing a gesture recognition procedure to generate a gesture recognition (GR) score representing a likelihood that the gesture signal corresponds to a first gesture command; and executing a first command in a vehicle corresponding to said first voice command and first gesture command when said voice recognition score exceeds a first threshold and said gesture recognition score exceeds a second threshold. | 1. A computer based method for executing a command in a vehicle, comprising the steps of: receiving a voice signal; receiving a gesture signal; performing a voice recognition procedure to generate a voice recognition (VR) score representing a likelihood that the voice signal corresponds to a first voice command; performing a gesture recognition procedure to generate a gesture recognition (GR) score representing a likelihood that the gesture signal corresponds to a first gesture command; and executing a first command in a vehicle corresponding to said first voice command and first gesture command when said voice recognition score exceeds a first threshold and said gesture recognition score exceeds a second threshold. 2. The method of claim 1 , further comprising the step of: generating a confirmation request, prior to executing said first command, when either (a) said voice recognition score exceeds a first threshold and said gesture recognition score does not exceed a second threshold or (b) said voice recognition score does not exceed a first threshold and said gesture recognition score exceeds a second threshold. | 0.568003 |
1. A method of formatting rich text data into a JavaScript Object Notation (JSON) array, the method comprising: creating a parent JSON object within the JSON array; identifying header information defined as a field of the rich text data and representing the header information in a root property of the parent JSON object; identifying a plurality of parts of the rich text data and, for each of the plurality of parts of the rich text data: via a processor, identifying content of the part, inserting into the parent JSON object a first respective child JSON object, and representing the content of the part in a content property of the respective child JSON object; and identifying header information of the part and representing the header information in a header property of the respective first child JSON object; wherein the children JSON objects are organized into a JSON array within the parent JSON object; and outputting the JSON array in a JSON file. | 1. A method of formatting rich text data into a JavaScript Object Notation (JSON) array, the method comprising: creating a parent JSON object within the JSON array; identifying header information defined as a field of the rich text data and representing the header information in a root property of the parent JSON object; identifying a plurality of parts of the rich text data and, for each of the plurality of parts of the rich text data: via a processor, identifying content of the part, inserting into the parent JSON object a first respective child JSON object, and representing the content of the part in a content property of the respective child JSON object; and identifying header information of the part and representing the header information in a header property of the respective first child JSON object; wherein the children JSON objects are organized into a JSON array within the parent JSON object; and outputting the JSON array in a JSON file. 5. The method of claim 1 , further comprising: assigning to a respective one of first children JSON objects a contentType property having a multipart/alternative attribute; converting alphanumeric text in the rich text data to plain text; and inserting into the respective one of the first children JSON objects a second child JSON object into which the plain text is placed, and assigning to the second child JSON object a contentType property having a text/plain attribute. | 0.592797 |
15. A system, comprising: a memory; and one or more processors coupled to the memory, wherein the memory comprises program instructions executable by the one or more processors to: based on reference data that includes pairs of information items and labels that each indicate whether a given pair of information items have a specific relationship, generate a first machine learning model for determining whether pairs of information items have said specific relationship; identify one or more false positive pairs, wherein a given false positive pair is a pair of information items that the first machine learning model indicates as having said specific relationship and which are labeled within the reference data as not having said specific relationship; generate an indication of one or more of the identified false positive pairs as candidates to be corrected within the reference data. | 15. A system, comprising: a memory; and one or more processors coupled to the memory, wherein the memory comprises program instructions executable by the one or more processors to: based on reference data that includes pairs of information items and labels that each indicate whether a given pair of information items have a specific relationship, generate a first machine learning model for determining whether pairs of information items have said specific relationship; identify one or more false positive pairs, wherein a given false positive pair is a pair of information items that the first machine learning model indicates as having said specific relationship and which are labeled within the reference data as not having said specific relationship; generate an indication of one or more of the identified false positive pairs as candidates to be corrected within the reference data. 20. The system of claim 15 , wherein the first machine learning model and the new machine model are different types of machine learning models. | 0.746743 |
1. A computer storage device storing computer-executable instructions that, when executed, cause one or more processors to perform operations comprising: receiving a plurality of media objects that are captured by an electronic device during a trip session; receiving one or more geolocations of the electronic device at periodic intervals during the trip session; determining a physical site visited during the trip session in part by identifying features depicted in the plurality of media objects; analyzing the one or more geolocations of the electronic device to determine a movement of the electronic device away from the physical site; auto-generating textual content for individual ones of the plurality of media objects that are captured by the electronic device during the trip session, the auto-generated textual content based at least in part on one or more pre-stored knowledge items that include information about the physical site; generating paragraph metadata objects for individual ones of the plurality of media objects, the paragraph metadata objects including at least one media object and the auto-generated textual content for the at least one media object; calculating paragraph weights based at least in part on the paragraph metadata objects, the paragraph weights associated with and representing a relative importance of individual ones of the paragraph metadata object to a user of the electronic device; and publishing a paragraph metadata object including the auto-generated textual content and the at least one media object as a blog entry for a place of interest that corresponds to the physical site when the associated paragraph weight is higher than paragraph weights associated with other paragraph metadata objects. | 1. A computer storage device storing computer-executable instructions that, when executed, cause one or more processors to perform operations comprising: receiving a plurality of media objects that are captured by an electronic device during a trip session; receiving one or more geolocations of the electronic device at periodic intervals during the trip session; determining a physical site visited during the trip session in part by identifying features depicted in the plurality of media objects; analyzing the one or more geolocations of the electronic device to determine a movement of the electronic device away from the physical site; auto-generating textual content for individual ones of the plurality of media objects that are captured by the electronic device during the trip session, the auto-generated textual content based at least in part on one or more pre-stored knowledge items that include information about the physical site; generating paragraph metadata objects for individual ones of the plurality of media objects, the paragraph metadata objects including at least one media object and the auto-generated textual content for the at least one media object; calculating paragraph weights based at least in part on the paragraph metadata objects, the paragraph weights associated with and representing a relative importance of individual ones of the paragraph metadata object to a user of the electronic device; and publishing a paragraph metadata object including the auto-generated textual content and the at least one media object as a blog entry for a place of interest that corresponds to the physical site when the associated paragraph weight is higher than paragraph weights associated with other paragraph metadata objects. 3. The computer storage device of claim 1 , the operations further comprising: clustering at least some photographs included in the plurality of media objects into a photograph cluster, the photograph cluster having a cluster center that is a mean of corresponding geolocations at which the at least some photographs are captured; identifying a plurality of candidate physical sites with nearest physical distances to the cluster center; and selecting a candidate physical site from the plurality of candidate physical sites as the place of interest when one or more first features depicted in a photograph of the photograph cluster match one or more second features depicted in a known photograph of the candidate physical site. | 0.5 |
12. A system comprising: one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising: receiving, by the one or more computers, audio data that describes an utterance; processing the audio data using a neural network that has been trained as an acoustic model, wherein the processing comprises: providing, as input to the neural network, input vectors having values describing the utterance, the values including values representing audio waveform features, wherein the audio waveform features are determined using a filterbank having parameters trained jointly with weights of the neural network, wherein the neural network has first memory blocks for time information and second memory blocks for frequency information, the first memory blocks being different from the second memory blocks; wherein the first memory blocks are time-LSTM blocks that each have a state, and wherein the second memory blocks are frequency-LSTM blocks that each have a state and a corresponding frequency step in a sequence of multiple frequency steps, wherein the states are determined for each of a sequence of multiple time steps; wherein, for each of at least some of the frequency-LSTM blocks, the frequency-LSTM block determines its state using the state of the time-LSTM block corresponding to the same frequency step at the previous time step; and wherein, for each of at least some of the time-LSTM blocks, the time-LSTM block determines its state using the state of the frequency-LSTM block corresponding to the same time step and the previous frequency step; receiving, as output of the neural network, one or more scores that each indicate a likelihood that a respective phonetic unit represents a portion of the utterance; determining, by the one or more computers, a transcription for the utterance based on the one or more scores; and providing, by the one or more computers, the determined transcription as output of an automated speech recognizer. | 12. A system comprising: one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising: receiving, by the one or more computers, audio data that describes an utterance; processing the audio data using a neural network that has been trained as an acoustic model, wherein the processing comprises: providing, as input to the neural network, input vectors having values describing the utterance, the values including values representing audio waveform features, wherein the audio waveform features are determined using a filterbank having parameters trained jointly with weights of the neural network, wherein the neural network has first memory blocks for time information and second memory blocks for frequency information, the first memory blocks being different from the second memory blocks; wherein the first memory blocks are time-LSTM blocks that each have a state, and wherein the second memory blocks are frequency-LSTM blocks that each have a state and a corresponding frequency step in a sequence of multiple frequency steps, wherein the states are determined for each of a sequence of multiple time steps; wherein, for each of at least some of the frequency-LSTM blocks, the frequency-LSTM block determines its state using the state of the time-LSTM block corresponding to the same frequency step at the previous time step; and wherein, for each of at least some of the time-LSTM blocks, the time-LSTM block determines its state using the state of the frequency-LSTM block corresponding to the same time step and the previous frequency step; receiving, as output of the neural network, one or more scores that each indicate a likelihood that a respective phonetic unit represents a portion of the utterance; determining, by the one or more computers, a transcription for the utterance based on the one or more scores; and providing, by the one or more computers, the determined transcription as output of an automated speech recognizer. 13. The system of claim 12 , wherein the neural network comprises a grid-LSTM module, a linear projection layer, one or more LSTM layers, and a deep neural network (DNN); wherein the grid-LSTM module includes the first memory blocks and the second memory blocks, and the grid-LSTM module provides output to the linear projection layer; wherein the linear projection layer provides output to the one or more LSTM layers; and wherein the one or more LSTM layers provide output to the DNN. | 0.5 |
12. An apparatus for identifying a candidate set of rules, the apparatus comprising: a memory; and at least one processor, coupled to the memory, operative to: identify one or more rules that receive information from one or more sensors to create a first candidate set of rules; evaluate each of said identified rules to identify an initial candidate set of rules, wherein a rule is added to said initial candidate set if a selected device is associated with sensors that are enabled to produce information for said rule being evaluated, if said selected device is enabled to locally provide input information required by said rule being evaluated, and if said rule being evaluated is associated with a group of devices and there are no correlation operators that consume events generated from other devices in said group of devices; and evaluate each rule in said initial candidate set to identify a final candidate set of rules, wherein any rule that receives information from said rule being evaluated is added to said initial candidate set to create said final candidate set if said selected device is enabled to locally satisfy input information required by said rule that receives information and if said rule that receives information is associated with a group of devices and there are no correlation operators that consume events generated from other devices in said group of devices. | 12. An apparatus for identifying a candidate set of rules, the apparatus comprising: a memory; and at least one processor, coupled to the memory, operative to: identify one or more rules that receive information from one or more sensors to create a first candidate set of rules; evaluate each of said identified rules to identify an initial candidate set of rules, wherein a rule is added to said initial candidate set if a selected device is associated with sensors that are enabled to produce information for said rule being evaluated, if said selected device is enabled to locally provide input information required by said rule being evaluated, and if said rule being evaluated is associated with a group of devices and there are no correlation operators that consume events generated from other devices in said group of devices; and evaluate each rule in said initial candidate set to identify a final candidate set of rules, wherein any rule that receives information from said rule being evaluated is added to said initial candidate set to create said final candidate set if said selected device is enabled to locally satisfy input information required by said rule that receives information and if said rule that receives information is associated with a group of devices and there are no correlation operators that consume events generated from other devices in said group of devices. 16. The apparatus of claim 12 , wherein said processor is further operative to replace one or more downloaded rules with one or more rules of said final candidate set if said one or more rules of said final candidate set have a higher priority than said one or more downloaded rules. | 0.617701 |
8. The computer program product of claim 7 , wherein automatically creating the at least one new column in the relational table comprises: for each read open tag, automatically determining a path from a root node corresponding to a first read open tag from the corresponding tag-language message to the currently read open tag; for each determined path: automatically designating the relational table in the relational database corresponding to the path, and storing a combination of at least one of (i) a first identifier for the currently read open tag corresponding to the path, the first identifier being stored in connection with information identifying a primary key column of the designated relational table corresponding to a primary key for the designated relational table, and (ii) a second identifier of a parent of the currently read open tag, the parent comprising the open tag immediately preceding the currently read open tag in the determined path, the second identifier being stored in connection with information identifying a foreign key column of the relational table corresponding to a foreign key of the relational table; for each read open tag having at least one of a value or an attribute value: designating at least one new data column different from the primary and foreign key columns of the relational table corresponding to the determined path for the read open tag, the at least one new data column corresponding to the at least one of the value or the attribute value, and storing information identifying the at least one new data column of the relational table in connection with the stored combination of the first identifier and the second identifier for the read open tag. | 8. The computer program product of claim 7 , wherein automatically creating the at least one new column in the relational table comprises: for each read open tag, automatically determining a path from a root node corresponding to a first read open tag from the corresponding tag-language message to the currently read open tag; for each determined path: automatically designating the relational table in the relational database corresponding to the path, and storing a combination of at least one of (i) a first identifier for the currently read open tag corresponding to the path, the first identifier being stored in connection with information identifying a primary key column of the designated relational table corresponding to a primary key for the designated relational table, and (ii) a second identifier of a parent of the currently read open tag, the parent comprising the open tag immediately preceding the currently read open tag in the determined path, the second identifier being stored in connection with information identifying a foreign key column of the relational table corresponding to a foreign key of the relational table; for each read open tag having at least one of a value or an attribute value: designating at least one new data column different from the primary and foreign key columns of the relational table corresponding to the determined path for the read open tag, the at least one new data column corresponding to the at least one of the value or the attribute value, and storing information identifying the at least one new data column of the relational table in connection with the stored combination of the first identifier and the second identifier for the read open tag. 11. The computer program product of claim 8 , wherein designating the relational table corresponding to the path comprises accessing a resource hashed on the path to determine if a relational table name has previously been designated for the path. | 0.759197 |
1. A computer implemented method for managing electronically delivered information channels, comprising: receiving, by a computing device having a channel personalization engine, from one of a first subscriber device associated with a subscriber and a second subscriber device associated with the subscriber, via a computer network and internet connection, a selection of non-textual category data entered by the subscriber via an interface of one of the first subscriber device and the second subscriber device, and a plurality of subscriber interest category data entered by the subscriber via the interface of one of the first subscriber device and the second subscriber device; identifying content associated with the non-textual category data and content associated with the subscriber interest category data; creating, by the channel personalization engine, a subscriber channel associated with the subscriber and configured to stream at least a portion of the content associated with the non-textual category data to the subscriber via the computer network and internet connection, and the subscriber channel configured to stream at least a portion of the content associated with the subscriber interest category data to the subscriber via the computer network and internet connection to at least one of the first subscriber device and the second subscriber device, the subscriber channel including the content associated with the non-textual category data and including the content associated with the subscriber interest category data, wherein the subscriber channel is a customized subscriber channel of the subscriber created by the channel personalization engine at least in part responsive to the non-textual category data entered by the subscriber and the plurality of subscriber interest category data entered by the subscriber; streaming, through the subscriber channel from the channel personalization engine via the computer network and internet connection to the first subscriber device, the portion of the content associated with the non-textual category data to the subscriber in a first format configured for display on the first subscriber device; streaming, through the subscriber channel from the channel personalization engine to the second subscriber device, the portion of the content associated with the subscriber interest category data to the subscriber in a second format configured for display on the second subscriber device during a time period of continuous transmission of the content associated with the non-textual category data from the channel personalization engine to the subscriber in the first format configured for display on the first subscriber device; receiving, by the channel personalization engine via the computer network and internet connection from a first user device, data entered into an interface of the first user device including at least one of second non-textual category data and user interest category data, wherein the first user device is a different device than the first subscriber device and the second subscriber device; determining, by the channel personalization engine, that a first user associated with the first user device has an interest in a portion of content of the subscriber channel, the determining at least in part responsive to the non-textual category data and the user interest category data entered by the first user via the first user device; streaming the portion of content of the subscriber channel to the first user via the first user device; and restricting the first user device's access of the subscriber channel to the portion of the content of the subscriber channel in which the first user associated with the first user device is determined to have the interest. | 1. A computer implemented method for managing electronically delivered information channels, comprising: receiving, by a computing device having a channel personalization engine, from one of a first subscriber device associated with a subscriber and a second subscriber device associated with the subscriber, via a computer network and internet connection, a selection of non-textual category data entered by the subscriber via an interface of one of the first subscriber device and the second subscriber device, and a plurality of subscriber interest category data entered by the subscriber via the interface of one of the first subscriber device and the second subscriber device; identifying content associated with the non-textual category data and content associated with the subscriber interest category data; creating, by the channel personalization engine, a subscriber channel associated with the subscriber and configured to stream at least a portion of the content associated with the non-textual category data to the subscriber via the computer network and internet connection, and the subscriber channel configured to stream at least a portion of the content associated with the subscriber interest category data to the subscriber via the computer network and internet connection to at least one of the first subscriber device and the second subscriber device, the subscriber channel including the content associated with the non-textual category data and including the content associated with the subscriber interest category data, wherein the subscriber channel is a customized subscriber channel of the subscriber created by the channel personalization engine at least in part responsive to the non-textual category data entered by the subscriber and the plurality of subscriber interest category data entered by the subscriber; streaming, through the subscriber channel from the channel personalization engine via the computer network and internet connection to the first subscriber device, the portion of the content associated with the non-textual category data to the subscriber in a first format configured for display on the first subscriber device; streaming, through the subscriber channel from the channel personalization engine to the second subscriber device, the portion of the content associated with the subscriber interest category data to the subscriber in a second format configured for display on the second subscriber device during a time period of continuous transmission of the content associated with the non-textual category data from the channel personalization engine to the subscriber in the first format configured for display on the first subscriber device; receiving, by the channel personalization engine via the computer network and internet connection from a first user device, data entered into an interface of the first user device including at least one of second non-textual category data and user interest category data, wherein the first user device is a different device than the first subscriber device and the second subscriber device; determining, by the channel personalization engine, that a first user associated with the first user device has an interest in a portion of content of the subscriber channel, the determining at least in part responsive to the non-textual category data and the user interest category data entered by the first user via the first user device; streaming the portion of content of the subscriber channel to the first user via the first user device; and restricting the first user device's access of the subscriber channel to the portion of the content of the subscriber channel in which the first user associated with the first user device is determined to have the interest. 3. The computer implemented method of claim 1 , comprising: providing, through the subscriber channel, the content associated with the non-textual category data to a plurality of users in the first format; and providing, through the subscriber channel, the content associated with the subscriber interest category data to the plurality of users in the second format. | 0.545745 |
13. A method of identifying pairs of similar vectors in a set of vectors, the method comprising: determining, using one or more computers, a partial similarity score for a vector x in a set of vectors and each other vector in the set of vectors, each partial similarity score representing a degree of similarity between features of the vector x and corresponding features of other vectors in the set of vectors; determining, using one or more computers, an upper bound, the upper bound being an estimate of the maximum similarity between non-processed features of the vector x and non-processed features of the other vectors, the non-processed features being features that have not been used to calculate the partial similarity scores; as long as the upper bound is greater than or equal to the similarity threshold, adding vectors to a candidate set of vectors and repeating the operations of determining a partial similarity score and determining an upper bound; when the upper bound is lower than the similarity threshold, determining partial similarity scores only for vectors in the candidate set of vectors; and identifying x and a vector y in the candidate set of vectors as similar vectors using the partial similarity score between x and y. | 13. A method of identifying pairs of similar vectors in a set of vectors, the method comprising: determining, using one or more computers, a partial similarity score for a vector x in a set of vectors and each other vector in the set of vectors, each partial similarity score representing a degree of similarity between features of the vector x and corresponding features of other vectors in the set of vectors; determining, using one or more computers, an upper bound, the upper bound being an estimate of the maximum similarity between non-processed features of the vector x and non-processed features of the other vectors, the non-processed features being features that have not been used to calculate the partial similarity scores; as long as the upper bound is greater than or equal to the similarity threshold, adding vectors to a candidate set of vectors and repeating the operations of determining a partial similarity score and determining an upper bound; when the upper bound is lower than the similarity threshold, determining partial similarity scores only for vectors in the candidate set of vectors; and identifying x and a vector y in the candidate set of vectors as similar vectors using the partial similarity score between x and y. 80. The method of claim 13 , in which each vector in the set of vectors represents a corresponding user in a community, and each feature of each vector represents the corresponding user's click-behavior with regard to a content item. | 0.582968 |
24. The apparatus of claim 15 , which comprises a user interface receiving a communication from a user, the communication defining the at least one reference word, the interrogation module identifying at least one rule associated with the at least one reference word, the at least one reference word and the at least one rule being stored in the reference database for interrogation when the incoming call is received. | 24. The apparatus of claim 15 , which comprises a user interface receiving a communication from a user, the communication defining the at least one reference word, the interrogation module identifying at least one rule associated with the at least one reference word, the at least one reference word and the at least one rule being stored in the reference database for interrogation when the incoming call is received. 25. The apparatus of claim 24 , wherein a plurality of configuration options are communicated to the user via an Interactive Voice Response (IVR) module, and responses received from the user being stored in the reference database as reference words and associated rules. | 0.800073 |
10. The method of claim 7 , wherein a subset of the entities and relationships between the entities are manipulated by a second layer overlaid over the first layer. | 10. The method of claim 7 , wherein a subset of the entities and relationships between the entities are manipulated by a second layer overlaid over the first layer. 11. The method of claim 10 , wherein the second layer is overlaid upon a selecting of text in an entity in the first layer. | 0.95058 |
1. A method of controlling access of a speaker to one of a service and a facility, the method comprising the steps of: (a) receiving first spoken utterances of the speaker, the first spoken utterances containing indicia of the speaker; (b) decoding the first spoken utterances; (c) accessing a database corresponding to the decoded first spoken utterances, the database containing information attributable to a speaker candidate having indicia substantially similar to the speaker; (d) querying the speaker with at least one question based on the information contained in the accessed database; (e) receiving second spoken utterances of the speaker, the second spoken utterances being representative of at least one answer to the at least one question; (f) decoding the second spoken utterances; (g) verifying the accuracy of the decoded answer against the information contained in the accessed database serving as the basis for the question; (h) taking a voice sample from the utterances of the speaker and processing the voice sample against an acoustic model attributable to the speaker candidate; (i) generating a score corresponding to the accuracy of the decoded answer and the closeness of the match between the voice sample and the model; and (j) comparing the score to a predetermined threshold value and if the score is one of substantially equivalent to and above the threshold value, then permitting speaker access to one of the service and the facility. | 1. A method of controlling access of a speaker to one of a service and a facility, the method comprising the steps of: (a) receiving first spoken utterances of the speaker, the first spoken utterances containing indicia of the speaker; (b) decoding the first spoken utterances; (c) accessing a database corresponding to the decoded first spoken utterances, the database containing information attributable to a speaker candidate having indicia substantially similar to the speaker; (d) querying the speaker with at least one question based on the information contained in the accessed database; (e) receiving second spoken utterances of the speaker, the second spoken utterances being representative of at least one answer to the at least one question; (f) decoding the second spoken utterances; (g) verifying the accuracy of the decoded answer against the information contained in the accessed database serving as the basis for the question; (h) taking a voice sample from the utterances of the speaker and processing the voice sample against an acoustic model attributable to the speaker candidate; (i) generating a score corresponding to the accuracy of the decoded answer and the closeness of the match between the voice sample and the model; and (j) comparing the score to a predetermined threshold value and if the score is one of substantially equivalent to and above the threshold value, then permitting speaker access to one of the service and the facility. 9. The method of claim 1, wherein at least a portion of the information in the database has static features. | 0.938279 |
2. The computer-implemented method of claim 1 , wherein determining the one or more query templates corresponding to different locales or languages includes determining probability distributions for the plurality of query templates. | 2. The computer-implemented method of claim 1 , wherein determining the one or more query templates corresponding to different locales or languages includes determining probability distributions for the plurality of query templates. 3. The computer-implemented method of claim 2 , wherein the probability distributions provide a likelihood that a certain query template is a correct interpretation for a given geographical search query and a given locale or language. | 0.925903 |
11. A method, comprising: receiving, at a device comprising a processor, a first media content item that comprises first low-level audio features and first video content having first low-level video features; determining that the first media content item belongs to a first genre of a plurality of genres based on first side channel information, wherein the first side channel information is based at least on the first low-level audio features and the first low-level video features; receiving a second media content item that comprises second audio content having second low-level audio features and second low-level video features; determining that the second media content item belongs to the first genre of the plurality of genres based on second side channel information, wherein the second side channel information is based at least on the second low-level audio features and the second low-level video features; generating a first fingerprint based on the first audio content; in response to determining that the first media content item belongs to the first genre, selecting, based on the first genre, a plurality of audio fingerprints that are associated with the first genre, wherein the selected plurality of audio fingerprints includes a second fingerprint that is based on the second audio content; determining a matching score between the first media content item and the second media content item by comparing the first audio fingerprint of the first audio content and the second audio fingerprint of the second audio content; determining a rate of false-positive matches between media content items that belong to the first genre; applying a weight to the matching score based on the rate of false-positive; determining, based on the weighted matching score, that the first audio content matches the second audio content; determining that the second audio content contains copyrighted content, wherein the copyrighted content is associated with a copyright owner; and notifying the copyright owner that the first media content item matches the second media content item. | 11. A method, comprising: receiving, at a device comprising a processor, a first media content item that comprises first low-level audio features and first video content having first low-level video features; determining that the first media content item belongs to a first genre of a plurality of genres based on first side channel information, wherein the first side channel information is based at least on the first low-level audio features and the first low-level video features; receiving a second media content item that comprises second audio content having second low-level audio features and second low-level video features; determining that the second media content item belongs to the first genre of the plurality of genres based on second side channel information, wherein the second side channel information is based at least on the second low-level audio features and the second low-level video features; generating a first fingerprint based on the first audio content; in response to determining that the first media content item belongs to the first genre, selecting, based on the first genre, a plurality of audio fingerprints that are associated with the first genre, wherein the selected plurality of audio fingerprints includes a second fingerprint that is based on the second audio content; determining a matching score between the first media content item and the second media content item by comparing the first audio fingerprint of the first audio content and the second audio fingerprint of the second audio content; determining a rate of false-positive matches between media content items that belong to the first genre; applying a weight to the matching score based on the rate of false-positive; determining, based on the weighted matching score, that the first audio content matches the second audio content; determining that the second audio content contains copyrighted content, wherein the copyrighted content is associated with a copyright owner; and notifying the copyright owner that the first media content item matches the second media content item. 14. The method of claim 11 , wherein the determining that the first media content item belongs to a first genre further comprises: implementing classifiers trained on at least one of: a first defined level of audio and a second defined level of video features of a plurality of media content items; or video metadata and audio metadata of the plurality of media content items. | 0.534879 |
10. The information processing device according to claim 1 , wherein the circuitry categorizes items known to the user and items unknown to the user. | 10. The information processing device according to claim 1 , wherein the circuitry categorizes items known to the user and items unknown to the user. 11. The information processing device according to claim 10 , wherein the circuitry performs a discrimination of whether an item is known to the user or unknown to the user, and performs learning of the discrimination. | 0.934076 |
1. A method of performing dependency mining, comprising: determining a block data source access sequence; splitting the block data source access sequence into a plurality of subsequences that represent a sequence of block accesses; determining an access pattern for each of the plurality of subsequences; constructing at least one search tree based on the access pattern of each of the plurality of subsequences; performing a search task using the at least one search tree; determining dependent blocks using the at least one search tree and based on block dependency criteria, wherein each access pattern is an activity vector that identifies occurrences of a particular block in the plurality of subsequences; outputting a dependency mining search result based on the search task and the block dependency criteria; and using the dependency mining search result in a storage management process in connection with at least one of the determined dependent blocks, wherein the block dependency criteria includes a second block being dependent on a first block in response to a confidence factor being greater than a predetermined threshold, the confidence factor corresponding to a number of occurrences where both the first block and second block are active divided by a number of occurrences where the first block is active, wherein subsets of the plurality of subsequences are grouped into a plurality of subset groups according to weighting characteristics of the block dependency criteria for each access pattern of the plurality of subsequences. | 1. A method of performing dependency mining, comprising: determining a block data source access sequence; splitting the block data source access sequence into a plurality of subsequences that represent a sequence of block accesses; determining an access pattern for each of the plurality of subsequences; constructing at least one search tree based on the access pattern of each of the plurality of subsequences; performing a search task using the at least one search tree; determining dependent blocks using the at least one search tree and based on block dependency criteria, wherein each access pattern is an activity vector that identifies occurrences of a particular block in the plurality of subsequences; outputting a dependency mining search result based on the search task and the block dependency criteria; and using the dependency mining search result in a storage management process in connection with at least one of the determined dependent blocks, wherein the block dependency criteria includes a second block being dependent on a first block in response to a confidence factor being greater than a predetermined threshold, the confidence factor corresponding to a number of occurrences where both the first block and second block are active divided by a number of occurrences where the first block is active, wherein subsets of the plurality of subsequences are grouped into a plurality of subset groups according to weighting characteristics of the block dependency criteria for each access pattern of the plurality of subsequences. 3. The method according to claim 1 , wherein more than one search tree is constructed, wherein the search task includes traversing the more than one search tree, and wherein the dependency mining search result is an aggregated result of traversing the more than one search tree. | 0.857798 |
4. An apparatus as in claim 3, wherein the base of each said word-building structure comprises a row of thirteen successive playing spaces, and each succeedingly higher row of said word-building structure has two less playing spaces than the row below it. | 4. An apparatus as in claim 3, wherein the base of each said word-building structure comprises a row of thirteen successive playing spaces, and each succeedingly higher row of said word-building structure has two less playing spaces than the row below it. 5. An apparatus as in claim 4, wherein said word-building structures each comprise seven rows of playing spaces. | 0.889212 |
9. The computer system of claim 8 , wherein the program instructions to identify comprise: program instructions to identify the particular experts as ones that have activity relating to the topic within a period of time from a current time at which an identification is made, wherein, for each one of the particular experts, the amount of activity of the particular expert is used to determine the likelihood of participation of the particular expert. | 9. The computer system of claim 8 , wherein the program instructions to identify comprise: program instructions to identify the particular experts as ones that have activity relating to the topic within a period of time from a current time at which an identification is made, wherein, for each one of the particular experts, the amount of activity of the particular expert is used to determine the likelihood of participation of the particular expert. 10. The computer system of claim 9 , wherein the activity is publishing papers, and wherein the published papers are related to the topic. | 0.953966 |
9. The method as recited in claim 1 , wherein the inferred response to the at least one of the one or more parent queries is automatically formed using a response technique selected by the user. | 9. The method as recited in claim 1 , wherein the inferred response to the at least one of the one or more parent queries is automatically formed using a response technique selected by the user. 11. The method as recited in claim 9 , wherein the response technique comprises forming the inferred response using a response from the subset of the one or more responses that indicates a most positive outcome. | 0.952358 |
15. A method, comprising: obtaining, by a device, text to be processed to identify a trend associated with a topic included in the text, the text including a plurality of text sections associated with the topic, the plurality of text sections being associated with a plurality of time frames; preparing, by the device, the text for processing, the text being prepared by standardizing the text for processing; determining, by the device, a respective context for the topic in each of the plurality of text sections, the respective context for the topic in each of the plurality of text sections including one or more terms that appear within a threshold distance of the topic; calculating, by the device, a first specificity score, associated with the topic for a first time frame, of the plurality of time frames, based on the respective context for the topic for one or more text sections, of the plurality of text sections, associated with the first time frame, the first specificity score indicating a degree to which text sections associated with the topic include specific information not included in other text sections; calculating, by the device, a second specificity score, associated with the topic for a second time frame, of the plurality of time frames, based on the respective context for the topic for one or more text sections, of the plurality of text sections, associated with the second time frame; calculating, by the device, a first trend score that represents a difference between the first specificity score and the second specificity score; calculating, by the device, a second trend score, the second trend score being associated with another topic; compare the first trend score and the second trend score; automatically identifying, by the device, the trend associated with the topic based on the first specificity score and the second specificity score thereby increasing speed and accuracy of identifying the trend; and providing, by the device, information that identifies the trend associated with the topic to permit an action to be taken based on the trend, the information being provided based on a result of comparing the first trend score and the second trend score. | 15. A method, comprising: obtaining, by a device, text to be processed to identify a trend associated with a topic included in the text, the text including a plurality of text sections associated with the topic, the plurality of text sections being associated with a plurality of time frames; preparing, by the device, the text for processing, the text being prepared by standardizing the text for processing; determining, by the device, a respective context for the topic in each of the plurality of text sections, the respective context for the topic in each of the plurality of text sections including one or more terms that appear within a threshold distance of the topic; calculating, by the device, a first specificity score, associated with the topic for a first time frame, of the plurality of time frames, based on the respective context for the topic for one or more text sections, of the plurality of text sections, associated with the first time frame, the first specificity score indicating a degree to which text sections associated with the topic include specific information not included in other text sections; calculating, by the device, a second specificity score, associated with the topic for a second time frame, of the plurality of time frames, based on the respective context for the topic for one or more text sections, of the plurality of text sections, associated with the second time frame; calculating, by the device, a first trend score that represents a difference between the first specificity score and the second specificity score; calculating, by the device, a second trend score, the second trend score being associated with another topic; compare the first trend score and the second trend score; automatically identifying, by the device, the trend associated with the topic based on the first specificity score and the second specificity score thereby increasing speed and accuracy of identifying the trend; and providing, by the device, information that identifies the trend associated with the topic to permit an action to be taken based on the trend, the information being provided based on a result of comparing the first trend score and the second trend score. 19. The method of claim 15 , where the text is natural language text. | 0.884742 |
9. The method of claim 7 , wherein replacing one of the plurality of edges with the one of the at least one second finite-state automaton based on the non-terminal symbol with which that edge is labeled comprises: modifying that second finite-state automaton into a new automaton that accepts the non-terminal symbol with which that edge is labeled; and substituting that edge with the new automaton. | 9. The method of claim 7 , wherein replacing one of the plurality of edges with the one of the at least one second finite-state automaton based on the non-terminal symbol with which that edge is labeled comprises: modifying that second finite-state automaton into a new automaton that accepts the non-terminal symbol with which that edge is labeled; and substituting that edge with the new automaton. 10. The method of claim 9 , wherein modifying that second finite-state automaton into a new automaton that accepts the non-terminal with which that edge is labeled comprises modifying that second finite-state automaton into a new automaton that accepts the non-terminal with which that edge is labeled based on a current status of the third finite-state transducer. | 0.70173 |
13. The computer readable storage device of claim 1 , wherein said one or more declarative lattice structures comprises a call function structure having one or more sites which act as input/output parameters. | 13. The computer readable storage device of claim 1 , wherein said one or more declarative lattice structures comprises a call function structure having one or more sites which act as input/output parameters. 15. The computer readable storage device of claim 13 , wherein said call function structure is called by one of said one or more declarative lattice structures. | 0.930597 |
1. A method comprising: receiving a string from an application at a first computing device; generating a first plurality of string predictions based on the received string by the first computing device, wherein each string prediction comprises a string and a confidence value and each string comprises a phrase that has been previously entered in response to the received string; providing one or more of the strings of the first plurality of string predictions according to the associated confidence values by the first computing device; receiving an indication of selection of one of the provided one or more strings by the first computing device; and in response to the indication of selection, providing the selected string as an input to the application by the first computing device. | 1. A method comprising: receiving a string from an application at a first computing device; generating a first plurality of string predictions based on the received string by the first computing device, wherein each string prediction comprises a string and a confidence value and each string comprises a phrase that has been previously entered in response to the received string; providing one or more of the strings of the first plurality of string predictions according to the associated confidence values by the first computing device; receiving an indication of selection of one of the provided one or more strings by the first computing device; and in response to the indication of selection, providing the selected string as an input to the application by the first computing device. 7. The method of claim 1 , further comprising: receiving a character; in response to the received character, generating a second plurality of string predictions based on the received string and the received character; and providing one or more of the strings of the second plurality of string predictions. | 0.52081 |
18. A cloud-based crowd-sourced query system, comprising: a general purpose computing device; and a computer program comprising program modules executable by the computing device, wherein the computing device is directed by the program modules of the computer program to: receive a crowd-sourced query comprising two or more branches, each branch associated with a corresponding set of worker qualifications; apply one or more automated optimizations to the crowd-sourced query to construct a reformulated query, the reformulated query having a reduced complexity relative to the crowd-sourced query; reduce one or more of expected completion time and expected cost associated with the reformulated query by associating the reformulated query with a matching optimized execution process selected from a plurality of predefined execution processes; provide the reformulated query and matching optimized execution process as an optimized version of the crowd-sourced query for execution via a cloud-based crowd-sourcing backend; during execution of the optimized version of the crowd-sourced query, improve query efficiency by automatically changing the matching optimized execution process to a different one of the predefined execution processes and dynamically changing the reformulated query to correspond to the changed execution process in response to collected runtime statistics relating to execution of the optimized version of the crowd-sourced query; and further during execution, presenting the changed reformulated query and changed matching optimized execution process for continued execution via the cloud-based crowd-sourcing backend. | 18. A cloud-based crowd-sourced query system, comprising: a general purpose computing device; and a computer program comprising program modules executable by the computing device, wherein the computing device is directed by the program modules of the computer program to: receive a crowd-sourced query comprising two or more branches, each branch associated with a corresponding set of worker qualifications; apply one or more automated optimizations to the crowd-sourced query to construct a reformulated query, the reformulated query having a reduced complexity relative to the crowd-sourced query; reduce one or more of expected completion time and expected cost associated with the reformulated query by associating the reformulated query with a matching optimized execution process selected from a plurality of predefined execution processes; provide the reformulated query and matching optimized execution process as an optimized version of the crowd-sourced query for execution via a cloud-based crowd-sourcing backend; during execution of the optimized version of the crowd-sourced query, improve query efficiency by automatically changing the matching optimized execution process to a different one of the predefined execution processes and dynamically changing the reformulated query to correspond to the changed execution process in response to collected runtime statistics relating to execution of the optimized version of the crowd-sourced query; and further during execution, presenting the changed reformulated query and changed matching optimized execution process for continued execution via the cloud-based crowd-sourcing backend. 19. The cloud-based crowd-sourced query system of claim 18 wherein the crowd-sourced query is formatted as a multi-layer structure, and wherein the automated optimizations include any combination of query flattening, query splitting, and common subexpression elimination. | 0.616959 |
2. The method of claim 1 , further comprising adding a sequence identifier to each of the first plurality of substrings to preserve the order of the first plurality of substrings. | 2. The method of claim 1 , further comprising adding a sequence identifier to each of the first plurality of substrings to preserve the order of the first plurality of substrings. 4. The method of claim 2 , further comprising adding a sequence identifier to each of the second plurality of substrings to preserve the order of the second plurality of substrings. | 0.935996 |
1. A document trimming apparatus for trimming documents having edges comprising: a table having sides, a top surface with a depression located substantially between said sides of said table and a slot located within said depression and extending through said table; a document cutting mechanism, operatively connected to the table, including a clamp holding a document while the document is being cut and a knife located above said table and defining a cutting line for cutting the document; a document positioning mechanism including a transport holder having a support member securely holding the document during transport, said support member being located in said depression, said document positioning mechanism further including a carrier movably supporting said transport holder for translational movement in said slot and translational and rotational movement of said support member in said depression relative to said document cutting mechanism, said document position mechanism positioning the document at predetermined positions proximate the cutting line. | 1. A document trimming apparatus for trimming documents having edges comprising: a table having sides, a top surface with a depression located substantially between said sides of said table and a slot located within said depression and extending through said table; a document cutting mechanism, operatively connected to the table, including a clamp holding a document while the document is being cut and a knife located above said table and defining a cutting line for cutting the document; a document positioning mechanism including a transport holder having a support member securely holding the document during transport, said support member being located in said depression, said document positioning mechanism further including a carrier movably supporting said transport holder for translational movement in said slot and translational and rotational movement of said support member in said depression relative to said document cutting mechanism, said document position mechanism positioning the document at predetermined positions proximate the cutting line. 5. A document trimming apparatus as defined in claim 1 wherein said document positioning mechanism includes an alignment block located at one side of said transport holder for locating the document when the document is initially placed in the document positioning mechanism, said alignment block being retractable. | 0.576117 |
1. A computer implemented method for accessing information from a plurality of searchable information sources, comprising: analyzing a search query to determine subject matter of the query; and selecting a subset of information sources from the plurality of information sources based upon the subject matter of the query; wherein the subject matter of the query is derived by comparing the search query against a knowledge-base, the knowledge base including a taxonomy of subject matters and a set of terms for at least some of the respective subject matters, the set of terms representing information likely to be found in the respective subject matters, and wherein comparing includes comparing at least portions of the search query against the sets of terms in the knowledge base to determine the respective subject matters of matching terms. | 1. A computer implemented method for accessing information from a plurality of searchable information sources, comprising: analyzing a search query to determine subject matter of the query; and selecting a subset of information sources from the plurality of information sources based upon the subject matter of the query; wherein the subject matter of the query is derived by comparing the search query against a knowledge-base, the knowledge base including a taxonomy of subject matters and a set of terms for at least some of the respective subject matters, the set of terms representing information likely to be found in the respective subject matters, and wherein comparing includes comparing at least portions of the search query against the sets of terms in the knowledge base to determine the respective subject matters of matching terms. 3. The computer implemented method of claim 1 , further comprising: comparing at least a portion of the search query against a plurality of entity lists; each entity list includes a list of phrases, each of the phrases corresponding to one or more subject matters; and wherein comparing includes matching a phrase in an entity list against at least a portion of the search query, and upon a match, returning subject matter corresponding to the match in the entity list. | 0.5 |
1. A method for transmitting at least one real-time customized notification alert to an electronic communication device of at least one user characterized by enabling a software-based platform to process a profile data and a contextual data associated with the at least one user to generate the customized notification, the method comprising computer-implemented steps of: acquiring the profile data for the user, wherein the profile data comprises user specific contextual information, demographic profile and a list of event types for which a notification alert in a specific format is desired; pre-processing, filtering, extending and on-demand loading or unloading structured background knowledge comprising geographical information and changing information about a plurality of users in order to derive a refined structured background knowledge relevant to the profile data for the at least one user using a set of one or more rules or one or more queries registered through an application program module communicatively coupled to the electronic communication device; receiving at least one raw event feed from one or more heterogeneous sources in assorted data formats, wherein the heterogeneous sources are a part of sensor network communicating with the electronic communication device though a communication network; determining a raw event data and a raw context data from the received raw event feed by using a context mapping database storing the user specific contextual information, wherein the received raw event feed is of at least one data format selected from a group comprising of text, image, video, audio, multimedia and combinations thereof, and wherein the raw event data is selected from a group comprising of traffic jams, robbery, flood, climate updates, traffic accident, road blockage, criminal activity, mutually affecting event among multiple users and combinations thereof; converting the raw event data and the raw context data into a structured knowledge format; loading a combined structured knowledge comprising the refined structured background knowledge, the structured raw event data and the structured raw context data into a stream reasoning module; applying stream reasoning by querying the combined structured knowledge to determine if the received raw event feed is relevant to the acquired profile data and the user-specific contextual information for the user; and transmitting the notification alert on the electronic communication device of the user in the specified format if the received raw event feed is identified as relevant by the applied stream reasoning. | 1. A method for transmitting at least one real-time customized notification alert to an electronic communication device of at least one user characterized by enabling a software-based platform to process a profile data and a contextual data associated with the at least one user to generate the customized notification, the method comprising computer-implemented steps of: acquiring the profile data for the user, wherein the profile data comprises user specific contextual information, demographic profile and a list of event types for which a notification alert in a specific format is desired; pre-processing, filtering, extending and on-demand loading or unloading structured background knowledge comprising geographical information and changing information about a plurality of users in order to derive a refined structured background knowledge relevant to the profile data for the at least one user using a set of one or more rules or one or more queries registered through an application program module communicatively coupled to the electronic communication device; receiving at least one raw event feed from one or more heterogeneous sources in assorted data formats, wherein the heterogeneous sources are a part of sensor network communicating with the electronic communication device though a communication network; determining a raw event data and a raw context data from the received raw event feed by using a context mapping database storing the user specific contextual information, wherein the received raw event feed is of at least one data format selected from a group comprising of text, image, video, audio, multimedia and combinations thereof, and wherein the raw event data is selected from a group comprising of traffic jams, robbery, flood, climate updates, traffic accident, road blockage, criminal activity, mutually affecting event among multiple users and combinations thereof; converting the raw event data and the raw context data into a structured knowledge format; loading a combined structured knowledge comprising the refined structured background knowledge, the structured raw event data and the structured raw context data into a stream reasoning module; applying stream reasoning by querying the combined structured knowledge to determine if the received raw event feed is relevant to the acquired profile data and the user-specific contextual information for the user; and transmitting the notification alert on the electronic communication device of the user in the specified format if the received raw event feed is identified as relevant by the applied stream reasoning. 6. The method of claim 1 , wherein the filtering step comprises selecting at least one knowledge dataset from the background knowledge that is relevant to data needed for generating relevant alerts for the user. | 0.931789 |
15. One or more machine-readable storage devices having stored therein a program product, which, when executed by a set of one or more processors, causes the set of one or more processors to perform a method comprising: for an expected input string comprising a plurality of expected string segments comprising a first expected string segment and a second expected string segment, receiving a first speech segment for the first expected string segment and a second speech segment for the second expected string segment, wherein the first speech segment is different from the second speech segment; performing speech recognition separately on the first speech segment and the second speech segment, wherein said performing speech recognition comprises generating a first segment n-best list and a second segment n-best list, the first segment n-best list comprising n highest confidence score results of said speech recognition on the first speech segment, and the second segment n-best list comprising n highest confidence score results of said speech recognition on the second speech segment, where n comprises at least one integer; generating a global n-best list corresponding to said expected input string, wherein the global n-best list comprises a plurality of results each generated at least in part by combining a result from said first segment n-best list with a result from said second segment n-best list; and determining a final global speech recognition result corresponding to said expected input string, wherein said determining said final global speech recognition result comprises pruning results of said global n-best list utilizing a pruning criterion. | 15. One or more machine-readable storage devices having stored therein a program product, which, when executed by a set of one or more processors, causes the set of one or more processors to perform a method comprising: for an expected input string comprising a plurality of expected string segments comprising a first expected string segment and a second expected string segment, receiving a first speech segment for the first expected string segment and a second speech segment for the second expected string segment, wherein the first speech segment is different from the second speech segment; performing speech recognition separately on the first speech segment and the second speech segment, wherein said performing speech recognition comprises generating a first segment n-best list and a second segment n-best list, the first segment n-best list comprising n highest confidence score results of said speech recognition on the first speech segment, and the second segment n-best list comprising n highest confidence score results of said speech recognition on the second speech segment, where n comprises at least one integer; generating a global n-best list corresponding to said expected input string, wherein the global n-best list comprises a plurality of results each generated at least in part by combining a result from said first segment n-best list with a result from said second segment n-best list; and determining a final global speech recognition result corresponding to said expected input string, wherein said determining said final global speech recognition result comprises pruning results of said global n-best list utilizing a pruning criterion. 16. The one or more machine-readable storage devices according to claim 15 , wherein generating a global n-best list comprises: generating a first result of the global n-best list by concatenating a first result of the first segment n-best list with a second result of the second segment n-best list; and generating a second result of the global n-best list by concatenating the first result of the first segment n-best list with a third result of the second segment n-best list. | 0.5 |
17. A system for generating a mixed-initiative dialog to obtain information for a set of information slots, the apparatus comprising: a processor operable to: select a subset of slots from the set of information slots, dependent upon a set of unfilled slots for which information is to be obtained in a current dialog cycle; construct a composite grammar dependent upon the selected subset of slots, comprising: selecting from specified grammar composition rules those rules that apply to the selected subset of slots; forming a slot grammar for each slot in the selected subset of slots using at least one corresponding rule in the selected grammar composition rules; and combining the slot grammars using the selected grammar composition rules to produce the composite grammar, said composite grammar being applicable to permutations of the selected subset of slots; generate prompt dependent upon the selected subset of slots; compare a user response to the composite grammar; and determine, dependent upon the comparing step, if the response provides relevant information for the set of unfilled slots; an output device adapted to present the prompt; and an input device for receiving the user response to the prompt. | 17. A system for generating a mixed-initiative dialog to obtain information for a set of information slots, the apparatus comprising: a processor operable to: select a subset of slots from the set of information slots, dependent upon a set of unfilled slots for which information is to be obtained in a current dialog cycle; construct a composite grammar dependent upon the selected subset of slots, comprising: selecting from specified grammar composition rules those rules that apply to the selected subset of slots; forming a slot grammar for each slot in the selected subset of slots using at least one corresponding rule in the selected grammar composition rules; and combining the slot grammars using the selected grammar composition rules to produce the composite grammar, said composite grammar being applicable to permutations of the selected subset of slots; generate prompt dependent upon the selected subset of slots; compare a user response to the composite grammar; and determine, dependent upon the comparing step, if the response provides relevant information for the set of unfilled slots; an output device adapted to present the prompt; and an input device for receiving the user response to the prompt. 19. The system of claim 17 , all the limitations of which are incorporated herein by reference, wherein the processor is further operable to determine if the execution flow of the current dialog cycle is complete; and if said flow is not complete, to perform one of the steps of disambiguating the user response, confirming the user response, and repeating the selecting, constructing, generating, receiving, comparing and determining steps for a subsequent subset of slots. | 0.599515 |
9. A program storage device readable by a machine, embodying a program of instructions for generating a ranked list of keywords, the instructions when executed by the machine perform a method, the method comprising: at a computer processor, maintaining a database recording a weighted relationship between keywords; receiving an initial keyword from a user; obtaining a stored set of web page listings previously associated with the initial keyword obtaining a set of keywords comprising, for each web page listing in the set of web page listings, keywords previously associated with the web page listings; ranking each keyword in the set of keywords based upon the weighted relationship between the initial keyword search and each keyword, thereby forming a ranked list of keywords; presenting the ranked list of keywords to the user; receiving, by the computer processor, an indication of a selected keyword that was selected by the user from the ranked list of keywords in response to the presenting; and responsive to receiving the indication of the selected keyword, updating the database to increase the weighted relationship between the initial keyword and the selected keyword. | 9. A program storage device readable by a machine, embodying a program of instructions for generating a ranked list of keywords, the instructions when executed by the machine perform a method, the method comprising: at a computer processor, maintaining a database recording a weighted relationship between keywords; receiving an initial keyword from a user; obtaining a stored set of web page listings previously associated with the initial keyword obtaining a set of keywords comprising, for each web page listing in the set of web page listings, keywords previously associated with the web page listings; ranking each keyword in the set of keywords based upon the weighted relationship between the initial keyword search and each keyword, thereby forming a ranked list of keywords; presenting the ranked list of keywords to the user; receiving, by the computer processor, an indication of a selected keyword that was selected by the user from the ranked list of keywords in response to the presenting; and responsive to receiving the indication of the selected keyword, updating the database to increase the weighted relationship between the initial keyword and the selected keyword. 16. The program storage device of claim 9 wherein the updating of the database uses a history factor associated with each relationship. | 0.575646 |
1. A method of synchronizing text with audio in a multimedia file, wherein the multimedia file includes previously synchronized video and audio, wherein the multimedia file has a start time and a stop time that defines a timeline for the multimedia file, wherein the frames of the video and the corresponding audio are each associated with respective points in time along the timeline, comprising the steps of: receiving the multimedia file and parsing the audio therefrom, but maintaining the timeline synchronization between the video and the audio; receiving closed-captioned data associated with the multimedia file, wherein the closed-captioned data contains closed-captioned text, wherein each word of the closed-captioned text is associated with a corresponding word spoken in the audio, wherein each word of the closed-captioned text has a high degree of accuracy with the corresponding word spoken in the audio but a low correlation with the respective point in time along the timeline at which the corresponding word was spoken in the audio; using automated speech recognition (ASR) software, generating ASR text of the parsed audio, wherein each word of the ASR text is associated approximately with the corresponding words spoken in the audio, wherein each word of the ASR text has a lower degree of accuracy with the corresponding words spoken in the audio than the respective words of the closed-captioned text but a high correlation with the respective point in time along the timeline at which the corresponding word was spoken in the audio; thereafter, using N-gram analysis, comparing each word of the closed-captioned text with a plurality of words of the ASR text until a match is found; for each matched word from the closed-captioned text, associating therewith the respective point in time along the timeline of the matched word from the ASR text corresponding therewith, whereby each closed-captioned word is associated with a respective point on the timeline corresponding to the same point in time on the timeline in which the word is actually spoken in the audio and occurs within the video. | 1. A method of synchronizing text with audio in a multimedia file, wherein the multimedia file includes previously synchronized video and audio, wherein the multimedia file has a start time and a stop time that defines a timeline for the multimedia file, wherein the frames of the video and the corresponding audio are each associated with respective points in time along the timeline, comprising the steps of: receiving the multimedia file and parsing the audio therefrom, but maintaining the timeline synchronization between the video and the audio; receiving closed-captioned data associated with the multimedia file, wherein the closed-captioned data contains closed-captioned text, wherein each word of the closed-captioned text is associated with a corresponding word spoken in the audio, wherein each word of the closed-captioned text has a high degree of accuracy with the corresponding word spoken in the audio but a low correlation with the respective point in time along the timeline at which the corresponding word was spoken in the audio; using automated speech recognition (ASR) software, generating ASR text of the parsed audio, wherein each word of the ASR text is associated approximately with the corresponding words spoken in the audio, wherein each word of the ASR text has a lower degree of accuracy with the corresponding words spoken in the audio than the respective words of the closed-captioned text but a high correlation with the respective point in time along the timeline at which the corresponding word was spoken in the audio; thereafter, using N-gram analysis, comparing each word of the closed-captioned text with a plurality of words of the ASR text until a match is found; for each matched word from the closed-captioned text, associating therewith the respective point in time along the timeline of the matched word from the ASR text corresponding therewith, whereby each closed-captioned word is associated with a respective point on the timeline corresponding to the same point in time on the timeline in which the word is actually spoken in the audio and occurs within the video. 5. The method of claim 1 wherein, for any unmatched word in the closed captioned text, identifying the closest matched words in the closed captioned text on either side of the unmatched word along the timeline and then comparing the unmatched word with words of the ASR text between the two points on the timeline and selecting the most likely match or matches thereto. | 0.557827 |
1. A computer-implemented method of simulating interactive communication between a user and a human subject, comprising: assigning at least one phrase to a stored content sequence, wherein the content sequence comprises a content clip of the subject, the subject being a human recorded on video, the content clip including a contemporaneously-recorded head and mouth of the subject and contemporaneously-recorded audio of the subject, wherein the content clip is free of any superimposed facial features; parsing the at least one phrase to produce at least one phonetic clone; associating the at least one phonetic clone with the stored content sequence; creating a transition between the content clip and the second content sequence by frame-matching a frame of the stored content sequence, the content sequence including the human subject speaking, with a frame of a second content sequence, the frame-matching being performed with respect to the recorded video of the entire head and facial features of the human subject; receiving an utterance from the user; matching the utterance to the at least one phonetic clone; and in response to matching the utterance, displaying the stored content sequence associated with the at least one phonetic clone in succession with the second content sequence. | 1. A computer-implemented method of simulating interactive communication between a user and a human subject, comprising: assigning at least one phrase to a stored content sequence, wherein the content sequence comprises a content clip of the subject, the subject being a human recorded on video, the content clip including a contemporaneously-recorded head and mouth of the subject and contemporaneously-recorded audio of the subject, wherein the content clip is free of any superimposed facial features; parsing the at least one phrase to produce at least one phonetic clone; associating the at least one phonetic clone with the stored content sequence; creating a transition between the content clip and the second content sequence by frame-matching a frame of the stored content sequence, the content sequence including the human subject speaking, with a frame of a second content sequence, the frame-matching being performed with respect to the recorded video of the entire head and facial features of the human subject; receiving an utterance from the user; matching the utterance to the at least one phonetic clone; and in response to matching the utterance, displaying the stored content sequence associated with the at least one phonetic clone in succession with the second content sequence. 3. The method of claim 1 , wherein parsing the phrase to produce the at least one phonetic clone further comprises: performing a second partial parsing of the phrase to produce at least one second partially parsed phrase; sub-parsing the at least one second partially parsed phrase to produce at least one second sub-parsed phrase; and generating at least one phonetic clone from the at least one second partially parsed phrase. | 0.638137 |
19. A system comprising: data processing apparatus programmed to perform operations comprising: analyzing a recent search activity period of a user to determine a short-term category of interest of the user, wherein the analyzing comprises comparing queries submitted by the user during the recent search activity period with categories of selected search results that were selected by the user during the recent search activity period; obtaining a plurality of search results responsive to a query, each of the plurality of search results having a respective score; for each particular search result of a first plurality of the search results: calculating a category selection value for the particular search result, wherein the category selection value for the particular search result is based on a measure of a count of selections of the particular search result as a portion of combined selection counts for search results responsive to the query for a plurality of users for the short-term category of interest; calculating a general selection value for the particular search result, wherein the general selection value for the particular search result is based on a measure of a count of selections of the particular search result as a portion of combined selection counts for search results responsive to the query for a plurality of users for any category of interest; calculating a category relevance for the particular search result, wherein the respective category relevance for the particular search result is based on a difference between the respective category selection value for the search result and the respective general selection value for the particular search result; selecting a one or more search results of the first plurality of search results, each selected search result being selected based on the selected search result having a category relevance that exceeds a threshold; adjusting the respective score for each of the selected search results based on, at least, the category selection value for each selected search result and the general selection value for each selected search result; and ranking the search results of the plurality of search results according to the respective adjusted scores for the selected search results and the respective scores for search results that were not selected. | 19. A system comprising: data processing apparatus programmed to perform operations comprising: analyzing a recent search activity period of a user to determine a short-term category of interest of the user, wherein the analyzing comprises comparing queries submitted by the user during the recent search activity period with categories of selected search results that were selected by the user during the recent search activity period; obtaining a plurality of search results responsive to a query, each of the plurality of search results having a respective score; for each particular search result of a first plurality of the search results: calculating a category selection value for the particular search result, wherein the category selection value for the particular search result is based on a measure of a count of selections of the particular search result as a portion of combined selection counts for search results responsive to the query for a plurality of users for the short-term category of interest; calculating a general selection value for the particular search result, wherein the general selection value for the particular search result is based on a measure of a count of selections of the particular search result as a portion of combined selection counts for search results responsive to the query for a plurality of users for any category of interest; calculating a category relevance for the particular search result, wherein the respective category relevance for the particular search result is based on a difference between the respective category selection value for the search result and the respective general selection value for the particular search result; selecting a one or more search results of the first plurality of search results, each selected search result being selected based on the selected search result having a category relevance that exceeds a threshold; adjusting the respective score for each of the selected search results based on, at least, the category selection value for each selected search result and the general selection value for each selected search result; and ranking the search results of the plurality of search results according to the respective adjusted scores for the selected search results and the respective scores for search results that were not selected. 23. The system of claim 19 , wherein adjusting the respective score for a particular selected search result comprises combining the category relevance for the selected search result with an information retrieval score of the particular selected search result. | 0.618776 |
7. The method of claim 1 , further comprising: comparing, by said computing system, said first associated risk level score to a predetermined risk level threshold, wherein said comparing determines that said first associated risk level score exceeds said predetermined risk level threshold; generating, by said computing system, an alert associated with said first associated risk level score exceeding said predetermined risk level threshold; and presenting by said computing system, said alert. | 7. The method of claim 1 , further comprising: comparing, by said computing system, said first associated risk level score to a predetermined risk level threshold, wherein said comparing determines that said first associated risk level score exceeds said predetermined risk level threshold; generating, by said computing system, an alert associated with said first associated risk level score exceeding said predetermined risk level threshold; and presenting by said computing system, said alert. 10. The method of claim 7 , further comprising: modifying, by said computing system in response to said alert, a transponder code for a transponder associated with said aircraft. | 0.958759 |
1. A computer-implemented method for representing markup language document data in a searchable format, comprising: parsing a markup language document into a data stream, wherein the data stream includes: a plurality of fields in a predefined format having a symbol table for at least one of the fields, wherein offset values for fields of the data stream are calculated automatically and without requiring storage in the data stream, and optimized field sizes based on a maximum value of data within each field; and storing the data stream in data storage. | 1. A computer-implemented method for representing markup language document data in a searchable format, comprising: parsing a markup language document into a data stream, wherein the data stream includes: a plurality of fields in a predefined format having a symbol table for at least one of the fields, wherein offset values for fields of the data stream are calculated automatically and without requiring storage in the data stream, and optimized field sizes based on a maximum value of data within each field; and storing the data stream in data storage. 8. The computer-implemented method of claim 1 , wherein the data stream resulting from the parsing includes a plurality of index fields, based on contents of the markup language document, configured to facilitate retrieval of the markup language document data. | 0.60442 |
19. A computer storage medium encoded with a computer program, the program comprising instructions that when executed by one or more computers cause the one or more computers to perform operations comprising receiving a current search query submitted by a current user of a current user device to a search engine system; determining that the current search query is similar to a first previously submitted search query of a plurality of previously submitted search queries that have previously been submitted by the current user to the search engine system, wherein determining that the current search query is similar to the first previously submitted search query comprises determining that at least one first term from the current search query matches a corresponding term that appears in the first previously submitted search query; determining that a different second term satisfies the condition that: (i) the different second term appears in the first previously submitted search query that is similar to the current search query, (ii) the different second term does not appear in the current search query, and (iii) the different second term appears in at least a threshold number of other distinct search queries of the plurality of previously submitted search queries that have previously been submitted by the current user, wherein each other distinct search query is distinct from both the first previously submitted search query and the current search query; generating a revised search query by adding the different second term to the current search query; obtaining search results for the revised search query from a search engine; and providing the search results to the current user in a response to the current search query. | 19. A computer storage medium encoded with a computer program, the program comprising instructions that when executed by one or more computers cause the one or more computers to perform operations comprising receiving a current search query submitted by a current user of a current user device to a search engine system; determining that the current search query is similar to a first previously submitted search query of a plurality of previously submitted search queries that have previously been submitted by the current user to the search engine system, wherein determining that the current search query is similar to the first previously submitted search query comprises determining that at least one first term from the current search query matches a corresponding term that appears in the first previously submitted search query; determining that a different second term satisfies the condition that: (i) the different second term appears in the first previously submitted search query that is similar to the current search query, (ii) the different second term does not appear in the current search query, and (iii) the different second term appears in at least a threshold number of other distinct search queries of the plurality of previously submitted search queries that have previously been submitted by the current user, wherein each other distinct search query is distinct from both the first previously submitted search query and the current search query; generating a revised search query by adding the different second term to the current search query; obtaining search results for the revised search query from a search engine; and providing the search results to the current user in a response to the current search query. 21. The computer storage medium of claim 19 , wherein the plurality of previously submitted search queries are search queries submitted by the current user within a recent time window of submitting the current query. | 0.714976 |
21. An improvement according to claim 20 wherein said first control means is further programmable such that the processor may send a block of data from the processor to the memory, said first formatting means formatting the block of data as the block of data is sequentially transferred. | 21. An improvement according to claim 20 wherein said first control means is further programmable such that the processor may send a block of data from the processor to the memory, said first formatting means formatting the block of data as the block of data is sequentially transferred. 22. An improvement according to claim 21 wherein said first control means is further programmable such that the processor may receive a block of data from the memory, said first formatting means formatting the block of data as the block of data is sequentially transferred. | 0.881152 |
16. A system for performing key phrase detection comprising: a memory configured to store a start state based rejection model and a key phrase model associated with a predetermined key phrase; a digital signal processor coupled to the memory, the digital signal processor to update, at a current time instance, the start state based rejection model and the key phrase model based on scores of sub-phonetic units representative of received audio input, wherein the start state based rejection model includes a single rejection state having a plurality of rejection model self loops each associated with a particular score of the scores of sub-phonetic units, wherein the key phrase model includes a plurality of key phrase states interconnected by transitions therebetween, wherein the start state based rejection model and the key phrase model are connected by a first transition from the single rejection state to a first key phrase state of the plurality of key phrase states, and wherein to update the start state based rejection model and the key phrase model, the digital signal processor is to transition a score from a particular key phrase state of the plurality of key phrase states of the key phrase model to a next key phrase state of the plurality of key phrase states of the key phrase model, to transition the score from the particular key phrase state to the single rejection state of the start state based rejection model, and to generate a rejection likelihood score corresponding to the single rejection state of the start state based rejection model and a key phrase likelihood score corresponding to the key phrase model; and detect the predetermined key phrase in the received audio input based on the rejection likelihood score and the key phrase likelihood score; and provide a wake indicator or a command in response to the detected predetermined key phrase. | 16. A system for performing key phrase detection comprising: a memory configured to store a start state based rejection model and a key phrase model associated with a predetermined key phrase; a digital signal processor coupled to the memory, the digital signal processor to update, at a current time instance, the start state based rejection model and the key phrase model based on scores of sub-phonetic units representative of received audio input, wherein the start state based rejection model includes a single rejection state having a plurality of rejection model self loops each associated with a particular score of the scores of sub-phonetic units, wherein the key phrase model includes a plurality of key phrase states interconnected by transitions therebetween, wherein the start state based rejection model and the key phrase model are connected by a first transition from the single rejection state to a first key phrase state of the plurality of key phrase states, and wherein to update the start state based rejection model and the key phrase model, the digital signal processor is to transition a score from a particular key phrase state of the plurality of key phrase states of the key phrase model to a next key phrase state of the plurality of key phrase states of the key phrase model, to transition the score from the particular key phrase state to the single rejection state of the start state based rejection model, and to generate a rejection likelihood score corresponding to the single rejection state of the start state based rejection model and a key phrase likelihood score corresponding to the key phrase model; and detect the predetermined key phrase in the received audio input based on the rejection likelihood score and the key phrase likelihood score; and provide a wake indicator or a command in response to the detected predetermined key phrase. 20. The system of claim 16 , wherein the key phrase likelihood score comprises a minimum of a first likelihood score associated with a second key phrase state of the key phrase model and a second likelihood score associated with a third key phrase state of the key phrase model. | 0.608422 |
1. An information extraction system for extracting identification information for identifying a problem evoking expression which is an expression evoking a problematic situation, the information extraction system comprising: hardware including a processor; a solution request sentence set acquisition unit implemented at least by the hardware and for acquiring, a sentence set matching a positive example solution request pattern representing a positive example of a sentence including the problem evoking expression as a positive example solution request sentence set, from a corpus, and a sentence set matching a negative example solution request pattern representing an opposite request to the positive example solution request as a negative example solution request sentence set, from the corpus, and extracting the problem evoking expression from a sentence in the positive example solution request sentence set and storing the sentence in association with the problem evoking expression, and extracting the problem evoking expression from a sentence in the negative example solution request sentence set and storing the sentence in association with the problem evoking expression; and an identification information specification unit implemented at least by the hardware and for comparing, for each problem evoking expression, constituent elements of sentences included in the positive example solution request sentence set and the negative example solution request sentence set, and specifying a constituent element characterizing the positive example solution request sentence set and a constituent element characterizing the negative example solution request sentence set, respectively as positive example identification information for identifying a positive example of a sentence including the problem evoking expression and negative example identification information for identifying a negative example of a sentence including the problem evoking expression. | 1. An information extraction system for extracting identification information for identifying a problem evoking expression which is an expression evoking a problematic situation, the information extraction system comprising: hardware including a processor; a solution request sentence set acquisition unit implemented at least by the hardware and for acquiring, a sentence set matching a positive example solution request pattern representing a positive example of a sentence including the problem evoking expression as a positive example solution request sentence set, from a corpus, and a sentence set matching a negative example solution request pattern representing an opposite request to the positive example solution request as a negative example solution request sentence set, from the corpus, and extracting the problem evoking expression from a sentence in the positive example solution request sentence set and storing the sentence in association with the problem evoking expression, and extracting the problem evoking expression from a sentence in the negative example solution request sentence set and storing the sentence in association with the problem evoking expression; and an identification information specification unit implemented at least by the hardware and for comparing, for each problem evoking expression, constituent elements of sentences included in the positive example solution request sentence set and the negative example solution request sentence set, and specifying a constituent element characterizing the positive example solution request sentence set and a constituent element characterizing the negative example solution request sentence set, respectively as positive example identification information for identifying a positive example of a sentence including the problem evoking expression and negative example identification information for identifying a negative example of a sentence including the problem evoking expression. 2. The information extraction system according to claim 1 , wherein the solution request sentence set acquisition unit implemented at least by the hardware acquires the positive example solution request sentence set and the negative example solution request sentence set in association with the problem evoking expression, using the positive example solution request pattern representing the problem evoking expression as a regular expression and the negative example solution request pattern representing the problem evoking expression as a regular expression. | 0.593182 |
1. In a medical implant assembly having at least one polyaxial bone screw attached to a longitudinal connecting member, the bone screw having a receiver with a channel, the improvement wherein: a) at least a portion of the longitudinal connecting member is sized and shaped to be received in the receiver channel; and further comprising: b) a compression insert directly engaging both the longitudinal connecting member and a shank of the polyaxial bone screw, the insert having a base, a pair of opposed arms with outer receiver engaging portions, and the opposed arms defining a through channel with a lower connecting member seating surface; wherein c) the compression insert is top-loadable in the receiver in a first orientation, wherein, when in the first orientation, the compression insert through channel is substantially perpendicular to the receiver channel, and then rotated to a second orientation, such that the compression insert through channel is substantially parallel to the receiver channel and cooperating receiver portions snap into the receiver engaging portions. | 1. In a medical implant assembly having at least one polyaxial bone screw attached to a longitudinal connecting member, the bone screw having a receiver with a channel, the improvement wherein: a) at least a portion of the longitudinal connecting member is sized and shaped to be received in the receiver channel; and further comprising: b) a compression insert directly engaging both the longitudinal connecting member and a shank of the polyaxial bone screw, the insert having a base, a pair of opposed arms with outer receiver engaging portions, and the opposed arms defining a through channel with a lower connecting member seating surface; wherein c) the compression insert is top-loadable in the receiver in a first orientation, wherein, when in the first orientation, the compression insert through channel is substantially perpendicular to the receiver channel, and then rotated to a second orientation, such that the compression insert through channel is substantially parallel to the receiver channel and cooperating receiver portions snap into the receiver engaging portions. 2. The improvement of claim 1 further comprising: a) at least one inner recessed relief surface formed in the compression insert, the recessed relief surface receiving material flow of the longitudinal connecting member. | 0.606164 |
1. A method of identifying voiced phonemes of human speech in real time, comprising the steps of: (a) detecting the starting points of glottal pulses occurring in the enunciation of a voiced phoneme; (b) computing, for an interval beginning at a glottal pulse and ending before the next glottal pulse, an approximation of the frequency and decay rate of at least the most dominant frequency component of the speech signal between adjacent glottal pulses; and (c) generating an identification of said phoneme based on said computation. | 1. A method of identifying voiced phonemes of human speech in real time, comprising the steps of: (a) detecting the starting points of glottal pulses occurring in the enunciation of a voiced phoneme; (b) computing, for an interval beginning at a glottal pulse and ending before the next glottal pulse, an approximation of the frequency and decay rate of at least the most dominant frequency component of the speech signal between adjacent glottal pulses; and (c) generating an identification of said phoneme based on said computation. 2. The method of claim 1, further comprising the step of determining the presence within said interval of second-most dominant and third-most dominant frequency components of said speech signal above a predetermined threshold level, and computing an approximation of at least the frequencies of said second-most and third-most dominant frequency components; and also further comprising the step of computing the frequency ratios of said most dominant frequency component to said second-most and third-most dominant frequency components, respectively. | 0.766068 |
1. A computing device comprising: one or more processors; and a non-transitory, computer-readable medium storing programming that is executable by the one or more processors, the programming comprising instructions to: receive an input data set comprising a document; determine at least one focus in the input data set, wherein the focus is at least one of a grammatical part of speech or a functional descriptor, and wherein the focus is a portion of the input data set less than the input data set; form a term unit matrix from the input data set, the term unit matrix comprising a plurality of term units represented as a plurality of numeric integer values, wherein the term unit matrix is a substantially canonical representation of the input data set; filter the plurality of term units by removing one or more term units from the plurality of term units based on the focus; for term units that remain after filtering, form a group of remaining term units based on an underlying grammatical rule of the input data set, wherein for each term unit of the group of remaining term units, the underlying grammatical rule is numerically encoded in respective numeric integer values of the remaining term units; identify at least one root term unit of the group of remaining term units, the at least one root term unit having a plurality of tail term units associated therewith; search a data repository that is different from the input data set using the at least one root term unit and the plurality of tail term units; organize search results based on the focus indicating presence of the at least one root term unit; and display the organized search results. | 1. A computing device comprising: one or more processors; and a non-transitory, computer-readable medium storing programming that is executable by the one or more processors, the programming comprising instructions to: receive an input data set comprising a document; determine at least one focus in the input data set, wherein the focus is at least one of a grammatical part of speech or a functional descriptor, and wherein the focus is a portion of the input data set less than the input data set; form a term unit matrix from the input data set, the term unit matrix comprising a plurality of term units represented as a plurality of numeric integer values, wherein the term unit matrix is a substantially canonical representation of the input data set; filter the plurality of term units by removing one or more term units from the plurality of term units based on the focus; for term units that remain after filtering, form a group of remaining term units based on an underlying grammatical rule of the input data set, wherein for each term unit of the group of remaining term units, the underlying grammatical rule is numerically encoded in respective numeric integer values of the remaining term units; identify at least one root term unit of the group of remaining term units, the at least one root term unit having a plurality of tail term units associated therewith; search a data repository that is different from the input data set using the at least one root term unit and the plurality of tail term units; organize search results based on the focus indicating presence of the at least one root term unit; and display the organized search results. 9. The computing device of claim 1 , wherein the underlying grammatical rule of the input data set is based on a human language. | 0.630787 |
16. An information retrieval system comprising: at least one processor; a query engine implemented on the at least one processor that receives a query and retrieves a ranked list of items from an index of items; an index generator that generates the index of items at least on the basis of predicted responses of a plurality of target users to a plurality of the items, each of the predicted responses having been generated by: accessing a computer-implemented contacts service to identify a plurality of second users being associated with at least one target user of the plurality of target users by the contacts service; eliciting predicted responses of the at least one target user from the plurality of second users, a predicted response directed toward an item of the plurality of items and indicating an opinion of the item predicted to be held by the at least one target user; storing a second user weight for each of the plurality of second users; calculating a prediction of the at least one target user's response to the item by combining the predicted responses from the plurality of second users using second user weights of the plurality of second users to generate a weighted average of the predicted responses; and a prediction generator arranged to identify at least one item of the plurality of items and the at least one target user of the plurality of target users having a number of predicted responses below a threshold value and elicit, from at least one of the plurality of second users, a predicted response of the at least one target user of the plurality of target users to the at least one item of the plurality of items; wherein, in response to receiving an actual response from the at least one target user indicating an opinion of the item held by the at least one target user, the prediction of the at least one target user's response is provided to the at least one target user. | 16. An information retrieval system comprising: at least one processor; a query engine implemented on the at least one processor that receives a query and retrieves a ranked list of items from an index of items; an index generator that generates the index of items at least on the basis of predicted responses of a plurality of target users to a plurality of the items, each of the predicted responses having been generated by: accessing a computer-implemented contacts service to identify a plurality of second users being associated with at least one target user of the plurality of target users by the contacts service; eliciting predicted responses of the at least one target user from the plurality of second users, a predicted response directed toward an item of the plurality of items and indicating an opinion of the item predicted to be held by the at least one target user; storing a second user weight for each of the plurality of second users; calculating a prediction of the at least one target user's response to the item by combining the predicted responses from the plurality of second users using second user weights of the plurality of second users to generate a weighted average of the predicted responses; and a prediction generator arranged to identify at least one item of the plurality of items and the at least one target user of the plurality of target users having a number of predicted responses below a threshold value and elicit, from at least one of the plurality of second users, a predicted response of the at least one target user of the plurality of target users to the at least one item of the plurality of items; wherein, in response to receiving an actual response from the at least one target user indicating an opinion of the item held by the at least one target user, the prediction of the at least one target user's response is provided to the at least one target user. 17. The information retrieval system as claimed in claim 16 wherein the query engine retrieves annotations comprising the predicted responses associated with the ranked list of items. | 0.5 |
9. A system for presenting information to a user, the system comprising: a processor; and a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the system to perform the operations of: receiving one or more ambiguous characters via a reduced-entry keypad of a wireless phone, the one or more ambiguous characters received as a sequence of numbers input through the reduced-entry keypad, each respective ambiguous character being a number that represents one of at least two disambiguated letters; exchanging at least one of the ambiguous characters with a host by transmitting the sequence of numbers to the host across a wireless network, exchanging the at least one of the ambiguous characters including exchanging the sequence of numbers upon receiving an amount of numbers in the sequence that meets an initial predetermined threshold amount of numbers, and exchanging subsequently received numbers, received as part of the sequence of numbers; receiving, from the host, results that represent disambiguated terms corresponding to the ambiguous characters exchanged with the host; rendering the results in a display of the wireless phone in a manner that enables identification of which of the disambiguated terms will be used upon a received selection of a displayed result; rendering advertisements in the display of the wireless phone, the advertisements responsive to the received sequence of ambiguous characters; receiving, from the host, updated results that represent disambiguated terms corresponding to the subsequently received numbers exchanged with the host; rendering the updated results in the display of the wireless phone; and in response to receiving a selection of one of the disambiguated terms, displaying information corresponding to the selection. | 9. A system for presenting information to a user, the system comprising: a processor; and a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the system to perform the operations of: receiving one or more ambiguous characters via a reduced-entry keypad of a wireless phone, the one or more ambiguous characters received as a sequence of numbers input through the reduced-entry keypad, each respective ambiguous character being a number that represents one of at least two disambiguated letters; exchanging at least one of the ambiguous characters with a host by transmitting the sequence of numbers to the host across a wireless network, exchanging the at least one of the ambiguous characters including exchanging the sequence of numbers upon receiving an amount of numbers in the sequence that meets an initial predetermined threshold amount of numbers, and exchanging subsequently received numbers, received as part of the sequence of numbers; receiving, from the host, results that represent disambiguated terms corresponding to the ambiguous characters exchanged with the host; rendering the results in a display of the wireless phone in a manner that enables identification of which of the disambiguated terms will be used upon a received selection of a displayed result; rendering advertisements in the display of the wireless phone, the advertisements responsive to the received sequence of ambiguous characters; receiving, from the host, updated results that represent disambiguated terms corresponding to the subsequently received numbers exchanged with the host; rendering the updated results in the display of the wireless phone; and in response to receiving a selection of one of the disambiguated terms, displaying information corresponding to the selection. 10. The system of claim 9 , wherein exchanging the at least one of the ambiguous characters with the host includes: sending the subsequently received numbers to the host without resending previously sent numbers of the sequence of numbers and without receiving a manually entered completion input that indicates a completion of the sequence of numbers; in response to receiving the subsequently received numbers in the sequence of numbers, and without receiving the completion input, determining that no relevant results exist; and in response to determining that no relevant results exist, indicating that that no relevant results exist. | 0.536885 |
1. A computer-implemented method comprising: receiving, with a processor executing a sequence verifier, a plurality of messages exchanged between two or more participants, the sequence verifier being a single module that is distinct from the two or more participants, the plurality of messages comprising a first message sent from a first participant of the two or more participants to a second participant of the two or more participants and a second message sent from the second participant of the two or more participants to the first participant of the two or more participants, forming a sequence; and testing conformance of each of the plurality of messages, with the processor executing the sequence verifier, to a specific markup language standard without using a reference implementation of the markup language standard; wherein testing conformance compares the first message in the sequence of messages exchanged between the first participant and the second participant to a valid sequence in a protocol verification graph to test conformance of the first message, and compares the second message in the sequence of messages exchanged between the first participant and the second participant to a valid sequence in the protocol verification graph to test conformance of the second message, the testing conformance verifying that the first message and second message conform to an appropriate message type based on a standard schema and were exchanged in a correct order based on the protocol verification graph; wherein the protocol verification graph describes constraints on a message including message sequence constraints and message type constraints, the protocol verification graph being created by analyzing an Extensible Markup Language (XML) standards specification. | 1. A computer-implemented method comprising: receiving, with a processor executing a sequence verifier, a plurality of messages exchanged between two or more participants, the sequence verifier being a single module that is distinct from the two or more participants, the plurality of messages comprising a first message sent from a first participant of the two or more participants to a second participant of the two or more participants and a second message sent from the second participant of the two or more participants to the first participant of the two or more participants, forming a sequence; and testing conformance of each of the plurality of messages, with the processor executing the sequence verifier, to a specific markup language standard without using a reference implementation of the markup language standard; wherein testing conformance compares the first message in the sequence of messages exchanged between the first participant and the second participant to a valid sequence in a protocol verification graph to test conformance of the first message, and compares the second message in the sequence of messages exchanged between the first participant and the second participant to a valid sequence in the protocol verification graph to test conformance of the second message, the testing conformance verifying that the first message and second message conform to an appropriate message type based on a standard schema and were exchanged in a correct order based on the protocol verification graph; wherein the protocol verification graph describes constraints on a message including message sequence constraints and message type constraints, the protocol verification graph being created by analyzing an Extensible Markup Language (XML) standards specification. 16. The method of claim 1 , wherein the markup language standard is based on an extensible markup language (XML). | 0.586977 |
38. A method of predicting items of media content likely to appeal to a user, comprising: predicting a rating of an item based on features explicitly rated by a user; in the absence of explicitly rated features, predicting a rating of an item based on implicitly rated features, wherein an adaptive modeling algorithm generates the implicit feature ratings based on previous overall ratings of items made by the user; and allowing the user to correct a predicted rating for an item in order to obtain predicted item ratings more in line with what the user expects; wherein the method is performed by one or more computing devices. | 38. A method of predicting items of media content likely to appeal to a user, comprising: predicting a rating of an item based on features explicitly rated by a user; in the absence of explicitly rated features, predicting a rating of an item based on implicitly rated features, wherein an adaptive modeling algorithm generates the implicit feature ratings based on previous overall ratings of items made by the user; and allowing the user to correct a predicted rating for an item in order to obtain predicted item ratings more in line with what the user expects; wherein the method is performed by one or more computing devices. 45. The method of claim 38 , wherein an explicit rating overrides an implicit rating. | 0.833151 |
1. A computer-implemented system comprising: one or more processors; a recognition component that is executable by the one or more processors to receive a partial query input as voice signals of a user with at least one other sensed input type comprising text, image or audio; a classifier component, executable by the one or more processors, that processes the partial query input and infers in real-time multiple different search goals based on the partial query input by accessing one or more query databases that store query information and from which similar or matching character sets, terms or phrases are derived for generating at least one complete query, the classifier comprising a support vector machine to find a hyperspace in a space of possible inputs to distinguish triggering input events from non-triggering input events, the classifier explicitly trained using generic training data and implicitly trained by observing user behavior; a query formulation component, executable by the one or more processors, that generates the at least one complete query based on the multiple different search goals; and a search engine, executable by the one or more processors, that receives the at least one complete query, presents the at least one complete query to the user for editing, and processes the at least one complete query to return search results for each of the at least one complete query, the search results based on a confidence value output by the voice recognition component indicating the confidence of converted voice signals relative to the partial query input as voice signals of the user. | 1. A computer-implemented system comprising: one or more processors; a recognition component that is executable by the one or more processors to receive a partial query input as voice signals of a user with at least one other sensed input type comprising text, image or audio; a classifier component, executable by the one or more processors, that processes the partial query input and infers in real-time multiple different search goals based on the partial query input by accessing one or more query databases that store query information and from which similar or matching character sets, terms or phrases are derived for generating at least one complete query, the classifier comprising a support vector machine to find a hyperspace in a space of possible inputs to distinguish triggering input events from non-triggering input events, the classifier explicitly trained using generic training data and implicitly trained by observing user behavior; a query formulation component, executable by the one or more processors, that generates the at least one complete query based on the multiple different search goals; and a search engine, executable by the one or more processors, that receives the at least one complete query, presents the at least one complete query to the user for editing, and processes the at least one complete query to return search results for each of the at least one complete query, the search results based on a confidence value output by the voice recognition component indicating the confidence of converted voice signals relative to the partial query input as voice signals of the user. 4. The system of claim 1 , further comprising a context component that generates context information related to a software environment in which the user is currently active, the context information employed by the classifier to infer the multiple different search goals. | 0.504399 |
19. At least one computer readable medium comprising a plurality of instructions that in response to being executed on an automatic speech recognition computing device, causes the speech recognition computing device to: propagate tokens, by at least one processor, that are each associated with a sound from a recording of a person talking and that is at least part of a word, through the weighted finite state transducer (WFST) having words or word identifiers as output labels of the WFST, and comprising placing word sequences into a word lattice; generate, by at least one processor, a word history designation for individual tokens when a word is established at a token propagating along an arc with an output symbol, wherein the word history designation indicates a word sequence; and determine, by at least one processor, whether or not two or more tokens should be combined to form a single token in a state of the WFST by using, at least in part, the word history designations so that the recorded sounds are transformed into data that indicates recognition of an utterance by using, at least in part, the word history designations. | 19. At least one computer readable medium comprising a plurality of instructions that in response to being executed on an automatic speech recognition computing device, causes the speech recognition computing device to: propagate tokens, by at least one processor, that are each associated with a sound from a recording of a person talking and that is at least part of a word, through the weighted finite state transducer (WFST) having words or word identifiers as output labels of the WFST, and comprising placing word sequences into a word lattice; generate, by at least one processor, a word history designation for individual tokens when a word is established at a token propagating along an arc with an output symbol, wherein the word history designation indicates a word sequence; and determine, by at least one processor, whether or not two or more tokens should be combined to form a single token in a state of the WFST by using, at least in part, the word history designations so that the recorded sounds are transformed into data that indicates recognition of an utterance by using, at least in part, the word history designations. 25. The medium of claim 19 , wherein the instructions cause the computing device to recombine two or more tokens in the same state of the WFST when the word history designations of the tokens are the same and avoiding a recombination when the word history designations of two or more tokens are not the same; place the word established at an arc of the WFST with an output label into a word lattice as tokens are being propagated; perform an exception update of the word lattice by recombining multiple tokens into a single new active token when the word history designations of the multiple tokens are different, and comprising: placing word lattice node references of the multiple tokens into the new active token when the references are unique to other references from other ones of the multiple tokens, and maintaining the reference with the best score for the new active token when the reference is the same among more than one of the multiple tokens; assign a different value to individual words in a vocabulary of possible words to be used as output symbols of the WFST, and using multiple values corresponding to multiple words to determine the word history designation; and combine multiple final end tokens into a single utterance end token; wherein the designation is a hash tag formed by using a recursive hash function, and wherein the word history designation is different depending on the order of the words within the word sequence. | 0.517878 |
11. A means for creating optimized elements of a device as in claim 1 , wherein adjusting the characteristics of said elements of said device, in accordance with said fuzzy cognitive map, is done by altering the design of a part. | 11. A means for creating optimized elements of a device as in claim 1 , wherein adjusting the characteristics of said elements of said device, in accordance with said fuzzy cognitive map, is done by altering the design of a part. 13. A means for creating optimized elements of a device as in claim 11 , wherein adjusting the characteristics of said elements of said device, in accordance with said fuzzy cognitive map, is done by altering the design of a part in response to the fluid dynamics of said part. | 0.908979 |
12. A non-transitory computer-readable medium having sets of instructions stored thereon which, when executed by a computer, cause the computer to: provide a query template which includes one or more bind variables, wherein the one or more bind variables are typeless within the CEP environment; provide sets of parameters corresponding to the one or more bind variables; parse the query template to determine positions of the one or more bind variables; scan the provided sets of parameters to determine which of the sets of parameters are to be bound to the one or more bind variables; bind the one or more bind variables which are determined to be bound to the corresponding sets of parameters; insert arbitrary predicates into the query template, based on the one or more bind variables being typeless; substitute the bound one or more bind variables with the corresponding sets of parameters; based on the sets of parameters, generate a single parameterized query which is a template that provides possible values for the bound one of more bind variables; determine a placeholder occurring in the single parameterized query for processing an event stream; based on the single parameterized query, generate multiple customized queries and views which differ by at least only one variable, wherein the multiple customized queries and views are maintained by substituting the placeholder at runtime with parameters from the sets of parameters; instantiate a new query based on the query temple which includes the corresponding sets of parameters substituted for the one or more bind variables and injecting the new query into a CEP server; based on the new query, build a query execution plan; adding the query execution plan to a runtime environment as a continuous query; and execute the continuous query to process the event stream. | 12. A non-transitory computer-readable medium having sets of instructions stored thereon which, when executed by a computer, cause the computer to: provide a query template which includes one or more bind variables, wherein the one or more bind variables are typeless within the CEP environment; provide sets of parameters corresponding to the one or more bind variables; parse the query template to determine positions of the one or more bind variables; scan the provided sets of parameters to determine which of the sets of parameters are to be bound to the one or more bind variables; bind the one or more bind variables which are determined to be bound to the corresponding sets of parameters; insert arbitrary predicates into the query template, based on the one or more bind variables being typeless; substitute the bound one or more bind variables with the corresponding sets of parameters; based on the sets of parameters, generate a single parameterized query which is a template that provides possible values for the bound one of more bind variables; determine a placeholder occurring in the single parameterized query for processing an event stream; based on the single parameterized query, generate multiple customized queries and views which differ by at least only one variable, wherein the multiple customized queries and views are maintained by substituting the placeholder at runtime with parameters from the sets of parameters; instantiate a new query based on the query temple which includes the corresponding sets of parameters substituted for the one or more bind variables and injecting the new query into a CEP server; based on the new query, build a query execution plan; adding the query execution plan to a runtime environment as a continuous query; and execute the continuous query to process the event stream. 13. The non-transitory computer-readable medium of claim 12 , wherein the sets of instructions when further executed by the computer, cause the computer to determine that new sets of parameters are specified for binding. | 0.573556 |
10. A system, comprising: a processor; a computer readable memory in circuit communication with the processor; and a computer readable storage medium in circuit communication with the processor; wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby: uses a stored baseline JavaScript Object Notation (JSON) file to render data values of a baseline data set into a baseline graphic presentation that meets a boundary condition requirement for displaying the data values, wherein a local reverse proxy server that is disposed in a network communication structure between a front end computer device and a back end server, in response to a request from the front end computer device, retrieves the stored baseline JSON file from a local file system of the front end computer device, and retrieves the data values of the baseline data set from the back end server, and wherein the front end computer device renders the data values of the baseline set of data into the baseline graphic presentation; stores the baseline graphic presentation in the local file system; uses the stored baseline JSON file to render in the front end device data values of a second set of data of the back end server data warehouse data into a new graphic presentation; compares the stored baseline graphic presentation to the new graphic presentation on a pixel-by-pixel basis; and in response to the pixel-by-pixel comparing indicating a difference in display of values of the boundary condition requirement in the new graphic presentation relative to the baseline graphic presentation, revises the stored baseline JSON file into a revised baseline JSON file that is stored on the local file system of the front end computer device and that renders the data values of the second set of data into a revised new graphic presentation, wherein comparing the revised new graphic presentation on a pixel-by-pixel basis to the baseline graphic presentation does not indicate a difference in display of the values of the boundary condition requirement in the revised new graphic presentation relative to the baseline graphic presentation. | 10. A system, comprising: a processor; a computer readable memory in circuit communication with the processor; and a computer readable storage medium in circuit communication with the processor; wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby: uses a stored baseline JavaScript Object Notation (JSON) file to render data values of a baseline data set into a baseline graphic presentation that meets a boundary condition requirement for displaying the data values, wherein a local reverse proxy server that is disposed in a network communication structure between a front end computer device and a back end server, in response to a request from the front end computer device, retrieves the stored baseline JSON file from a local file system of the front end computer device, and retrieves the data values of the baseline data set from the back end server, and wherein the front end computer device renders the data values of the baseline set of data into the baseline graphic presentation; stores the baseline graphic presentation in the local file system; uses the stored baseline JSON file to render in the front end device data values of a second set of data of the back end server data warehouse data into a new graphic presentation; compares the stored baseline graphic presentation to the new graphic presentation on a pixel-by-pixel basis; and in response to the pixel-by-pixel comparing indicating a difference in display of values of the boundary condition requirement in the new graphic presentation relative to the baseline graphic presentation, revises the stored baseline JSON file into a revised baseline JSON file that is stored on the local file system of the front end computer device and that renders the data values of the second set of data into a revised new graphic presentation, wherein comparing the revised new graphic presentation on a pixel-by-pixel basis to the baseline graphic presentation does not indicate a difference in display of the values of the boundary condition requirement in the revised new graphic presentation relative to the baseline graphic presentation. 11. The system of claim 10 , wherein the processor executes the program instructions stored on the computer-readable storage medium via the computer readable memory and thereby further: creates an initial JSON file that causes the front end computer device to generate an initial front end graphic presentation of the data values of the baseline set of data of the back end data warehouse server that fails to meet the boundary condition requirement for displaying the data values; revises the initial JSON file into a revised JSON file in an iterative process until the revised JSON file is used by the front end computer device to render graphic presentation results that meet the boundary condition requirement; and stores the revised JSON file as the stored baseline JSON file that is stored in the local file system for retrieval to the local reverse proxy server in response to the request from the front end computer device. | 0.586439 |
6. A method for processing non-semantic visual input data to organize the non-semantic visual input data so that the non-semantic visual input data is readily accessible by both a human and a computing device, by describing the non-semantic visual input data semantically, comprising: (a) enabling a user to select and access portions of the non-semantic visual input data that have not yet been tagged with asserts, for tagging with asserts according to defined criteria; (b) selecting elements in the portions of the non-semantic visual input data that have been tagged with asserts, for further processing, wherein one type selected from a plurality of different types is assigned to each selected element, each type that is assigned being associated with a corresponding assert; (c) generating additional asserts that connect the asserts associated with each type of element, to properties of said element; and (d) for one or more types assigned to the selected elements, enabling one or more additional properties having their own tagonomy to be associated with the selected elements, wherein the asserts and properties associated with the non-semantic visual input data are in a form that enables the non-semantic visual input data to be readily accessed and queried by both a human and a computing device. | 6. A method for processing non-semantic visual input data to organize the non-semantic visual input data so that the non-semantic visual input data is readily accessible by both a human and a computing device, by describing the non-semantic visual input data semantically, comprising: (a) enabling a user to select and access portions of the non-semantic visual input data that have not yet been tagged with asserts, for tagging with asserts according to defined criteria; (b) selecting elements in the portions of the non-semantic visual input data that have been tagged with asserts, for further processing, wherein one type selected from a plurality of different types is assigned to each selected element, each type that is assigned being associated with a corresponding assert; (c) generating additional asserts that connect the asserts associated with each type of element, to properties of said element; and (d) for one or more types assigned to the selected elements, enabling one or more additional properties having their own tagonomy to be associated with the selected elements, wherein the asserts and properties associated with the non-semantic visual input data are in a form that enables the non-semantic visual input data to be readily accessed and queried by both a human and a computing device. 8. The method of claim 6 , further comprising the step of enabling a computing device to preselect elements in the non-semantic visual input data from among the portions of the non-semantic visual input data that have been tagged with asserts to indicate that further processing of the portions is appropriate, at least some of the elements that are preselected by the computing device being further processed to assign the type of the element. | 0.595396 |
10. A method to detect garbled closed captioning data, comprising: detecting closed captioning data in a video data stream; identifying and extracting individual words from the closed captioning data; determining a word boundary in the closed captioning data using a delimiter; storing a count of the total number of words in the closed captioning data in a memory based on the determined word boundary; storing a count of the total number of words having a desired word length or range of word lengths in the closed captioning data in the memory based on the determined word boundary; determining a percentage of words having the desired length or range of lengths in the closed captioning data as a ratio of the count of the number of words in the closed captioning data having the desired length or range of lengths to the count of the total number of words in the closed captioning data; and providing an alert when the determined percentage exceeds a predetermined threshold. | 10. A method to detect garbled closed captioning data, comprising: detecting closed captioning data in a video data stream; identifying and extracting individual words from the closed captioning data; determining a word boundary in the closed captioning data using a delimiter; storing a count of the total number of words in the closed captioning data in a memory based on the determined word boundary; storing a count of the total number of words having a desired word length or range of word lengths in the closed captioning data in the memory based on the determined word boundary; determining a percentage of words having the desired length or range of lengths in the closed captioning data as a ratio of the count of the number of words in the closed captioning data having the desired length or range of lengths to the count of the total number of words in the closed captioning data; and providing an alert when the determined percentage exceeds a predetermined threshold. 18. The method recited in claim 10 , further comprising detecting closed captioning data in MPEG-2 format or MPEG-4 format. | 0.546435 |
13. The article of claim 10 further storing instructions that, if executed, enable the processor-based system to determine a kernel based supervised classifier. | 13. The article of claim 10 further storing instructions that, if executed, enable the processor-based system to determine a kernel based supervised classifier. 14. The article of claim 13 further storing instructions that, if executed, enable a processor-based system to set the number of kernels to equal the number of dimensions in a data space plus one. | 0.826087 |
17. An apparatus that determines superwords and meaningful phases, comprising: a database of observed sequences of words, symbols and/or sounds containing a generated set of candidate phrases, the database having a perplexity value based on a language model; a superword selector that selects superwords, the superword selector incorporating the selected superwords from the database into the language model; and a meaningful phrase selector that selects meaningful phrases based on commonality measurements, the meaningful phrase selector incorporating the selected meaningful phrases from the database into the language model. | 17. An apparatus that determines superwords and meaningful phases, comprising: a database of observed sequences of words, symbols and/or sounds containing a generated set of candidate phrases, the database having a perplexity value based on a language model; a superword selector that selects superwords, the superword selector incorporating the selected superwords from the database into the language model; and a meaningful phrase selector that selects meaningful phrases based on commonality measurements, the meaningful phrase selector incorporating the selected meaningful phrases from the database into the language model. 21. The apparatus of claim 17, wherein the meaningful phrase selector selects a set of meaningful phrases based on salience and mutual information measurements. | 0.521678 |
10. A system comprising: a database source computer module configured to extract data associated with a plurality of co-occurring topics in a document corpus; and one or more computers comprising one or more processors configured to: identify, in the document corpus stored in the database source, an indication of a topic of interest; automatically extract from a document corpus, data associated with a plurality of co-occurring topics based on the topic of interest; extract a plurality of topic identifiers from the plurality of co-occurring topics in response to the extracting of the data associated with the plurality of co-occurring topics; create a periodic topic model comprising a set of one or more term vectors by comparing topic significance among the plurality of topic identifiers; periodically create new topic ID models using data content in the periodic topic model by identifying a similarity of topics, wherein the new topic ID models are stored in an in-memory database; and link data in the in-memory database for automated discovery of new topics. | 10. A system comprising: a database source computer module configured to extract data associated with a plurality of co-occurring topics in a document corpus; and one or more computers comprising one or more processors configured to: identify, in the document corpus stored in the database source, an indication of a topic of interest; automatically extract from a document corpus, data associated with a plurality of co-occurring topics based on the topic of interest; extract a plurality of topic identifiers from the plurality of co-occurring topics in response to the extracting of the data associated with the plurality of co-occurring topics; create a periodic topic model comprising a set of one or more term vectors by comparing topic significance among the plurality of topic identifiers; periodically create new topic ID models using data content in the periodic topic model by identifying a similarity of topics, wherein the new topic ID models are stored in an in-memory database; and link data in the in-memory database for automated discovery of new topics. 11. The system of claim 10 , wherein the one or more computers are further configured to determine a relationship of corresponding term vectors from the plurality of co-occurring topics where each co-occurring topic of the plurality of co-occurring topics containing one or more term vectors. | 0.5 |
1. A method performed by one or more processing devices, comprising: receiving, from a first user, information specifying one or more attributes of a story related to a second user; obtaining content items from a social network, with the content items comprising one or more content items in the social network related to the story and related to the second user and one or more content items in the social network unrelated to the story and related to the second user; filtering the obtained content items to include only one or more of the one or more content items related to the story and related to the second user, with a content item being related to the story when the content item satisfies one or more of the one or more attributes, and with the content item being related to the second user when a node representing the content item in a social graph of the social network is connected in the social graph to another node representing the second user; and generating, based on the filtered content, data for a graphical user interface that when rendered by a device used by the first user, comprises a visual representation of the story. | 1. A method performed by one or more processing devices, comprising: receiving, from a first user, information specifying one or more attributes of a story related to a second user; obtaining content items from a social network, with the content items comprising one or more content items in the social network related to the story and related to the second user and one or more content items in the social network unrelated to the story and related to the second user; filtering the obtained content items to include only one or more of the one or more content items related to the story and related to the second user, with a content item being related to the story when the content item satisfies one or more of the one or more attributes, and with the content item being related to the second user when a node representing the content item in a social graph of the social network is connected in the social graph to another node representing the second user; and generating, based on the filtered content, data for a graphical user interface that when rendered by a device used by the first user, comprises a visual representation of the story. 6. The method of claim 1 , wherein the one or more attributes comprise information identifying a subject of the story, the subject being a person who is different from the second user. | 0.734195 |
1. A method implemented by one or more computers comprising one or more processors, the method comprising: receiving an image query; receiving ranked image search results responsive the image query; generating, by the one or more processors, a hierarchical grouping of the image search results based on measures of similarity between images identified by the image search results; identifying, by the one or more processors, a respective canonical image for each of a plurality of groups in the hierarchical grouping based on a likelihood that users will interact with the canonical image when presented with multiple images from the group; and providing, in response to the image query and by the one or more processors, data that present a visual representation of the identified canonical images wherein canonical images for lower levels of the hierarchy are presented smaller than canonical images for higher levels of the hierarchy. | 1. A method implemented by one or more computers comprising one or more processors, the method comprising: receiving an image query; receiving ranked image search results responsive the image query; generating, by the one or more processors, a hierarchical grouping of the image search results based on measures of similarity between images identified by the image search results; identifying, by the one or more processors, a respective canonical image for each of a plurality of groups in the hierarchical grouping based on a likelihood that users will interact with the canonical image when presented with multiple images from the group; and providing, in response to the image query and by the one or more processors, data that present a visual representation of the identified canonical images wherein canonical images for lower levels of the hierarchy are presented smaller than canonical images for higher levels of the hierarchy. 6. The method of claim 1 , wherein the measures of similarity between the images is based on one or more of content-based signals, user behavior-based signals, or text-based signals. | 0.597478 |
1. A driving maneuver assisting apparatus comprising: a learning section configured to learn a driving-behavior pattern of a driver for a predetermined duration; a non-steady-state degree calculating section configured to calculate a non-steady-state degree by comparing a current driving-behavior pattern with the driving-behavior pattern learned by the learning section, wherein the non-steady-state degree represents how different the current driving-behavior pattern is from the driving-behavior pattern learned by the learning section; a learning level calculating section configured to calculate a learning level of the learning section; and a notifying section configured to notify the driver of maneuver assisting information for inducing the driving-behavior pattern learned by the learning section in accordance with the learning level calculated by the learning level calculating section, when the non-steady-state degree calculated by the non-steady-state degree calculating section exceeds a threshold value, wherein the notifying section is configured to provide contents of the maneuver assisting information in more detail as the learning level calculated by the learning level calculating section becomes higher, and wherein the learning level calculating section is configured to calculate the learning level of the learning section in accordance with a number of data points already acquired as the driving-behavior pattern of the driver by the learning section. | 1. A driving maneuver assisting apparatus comprising: a learning section configured to learn a driving-behavior pattern of a driver for a predetermined duration; a non-steady-state degree calculating section configured to calculate a non-steady-state degree by comparing a current driving-behavior pattern with the driving-behavior pattern learned by the learning section, wherein the non-steady-state degree represents how different the current driving-behavior pattern is from the driving-behavior pattern learned by the learning section; a learning level calculating section configured to calculate a learning level of the learning section; and a notifying section configured to notify the driver of maneuver assisting information for inducing the driving-behavior pattern learned by the learning section in accordance with the learning level calculated by the learning level calculating section, when the non-steady-state degree calculated by the non-steady-state degree calculating section exceeds a threshold value, wherein the notifying section is configured to provide contents of the maneuver assisting information in more detail as the learning level calculated by the learning level calculating section becomes higher, and wherein the learning level calculating section is configured to calculate the learning level of the learning section in accordance with a number of data points already acquired as the driving-behavior pattern of the driver by the learning section. 3. The driving maneuver assisting apparatus as claimed in claim 1 , wherein the notifying section is configured to vary the contents of the maneuver assisting information in accordance with the learning level calculated by the learning level calculating section. | 0.618497 |
6. The method of claim 1 wherein the identifying of a phrase includes selecting words with top scores, locating each selected word within the web pages as a part of a phrase, and extending each phrase by words proximate to the phrase based at least in part on the first score for said at least some of the words. | 6. The method of claim 1 wherein the identifying of a phrase includes selecting words with top scores, locating each selected word within the web pages as a part of a phrase, and extending each phrase by words proximate to the phrase based at least in part on the first score for said at least some of the words. 7. The method of claim 6 wherein a phrase is only extended by words that have an associated first score indicating relatedness to the item. | 0.90181 |
11. A computerized system comprising: one or more processors; and one or more computer storage media storing computer-usable instructions that, when used by the one or more processors, cause the one or more processors to: receive, via a displayed graphical user interface (GUI), an instruction to obtain a biometric signature from a remote computing device for association with a signature field of a displayed digital document, the received instruction including an electronic contact address associated with the remote computing device; send a remote signing request associated with the signature field of the displayed digital document to a remote server device, the sent remote signing request causing the remote server device to send, to the electronic contact address, a generated URI that immediately presents at least one option to receive the biometric signature when accessed; and place the biometric signature into a position that corresponds to the signature field of the digital document based on a receipt of the biometric signature from the remote server device. | 11. A computerized system comprising: one or more processors; and one or more computer storage media storing computer-usable instructions that, when used by the one or more processors, cause the one or more processors to: receive, via a displayed graphical user interface (GUI), an instruction to obtain a biometric signature from a remote computing device for association with a signature field of a displayed digital document, the received instruction including an electronic contact address associated with the remote computing device; send a remote signing request associated with the signature field of the displayed digital document to a remote server device, the sent remote signing request causing the remote server device to send, to the electronic contact address, a generated URI that immediately presents at least one option to receive the biometric signature when accessed; and place the biometric signature into a position that corresponds to the signature field of the digital document based on a receipt of the biometric signature from the remote server device. 12. The system of claim 11 , wherein the electronic contact address is one of a mobile device phone number, an email address, or a social media account identifier. | 0.591451 |
8. A character string correction system for use in an information processing system for analyzing a morpheme by comparing an input character string with a dictionary entry, comprising: dictionary storage means for storing a dictionary having entries of input characters to be compared with characters in the input character string; error pattern storage means for storing an error pattern prescribing a type of possible error in the input character string; retrieval means for searching the dictionary stored in said dictionary storage means using the error pattern stored in said error pattern storage means retrieving the dictionary entry corresponding to the input character string, outputting the retrieved dictionary entry as a candidate for a recognized word, and for generating a corresponding analysis path when a character in the input character string matches an entry of the input character in the dictionary; and memory means for storing an analysis path indicating a retrieval path from a first character to an intermediate character of the dictionary entry. | 8. A character string correction system for use in an information processing system for analyzing a morpheme by comparing an input character string with a dictionary entry, comprising: dictionary storage means for storing a dictionary having entries of input characters to be compared with characters in the input character string; error pattern storage means for storing an error pattern prescribing a type of possible error in the input character string; retrieval means for searching the dictionary stored in said dictionary storage means using the error pattern stored in said error pattern storage means retrieving the dictionary entry corresponding to the input character string, outputting the retrieved dictionary entry as a candidate for a recognized word, and for generating a corresponding analysis path when a character in the input character string matches an entry of the input character in the dictionary; and memory means for storing an analysis path indicating a retrieval path from a first character to an intermediate character of the dictionary entry. 10. The character string correction system according to claim 8, wherein when recognizing a morpheme as a result of searching the dictionary, said retrieval means stores information specifying the morpheme as being associated with the analysis path. | 0.762357 |
15. An electronic device for use in a distributed speech recognition system comprising the electronic device and a network device remote from the electronic device, the electronic device, comprising: at least one storage device configured to store one or more applications; an embedded speech recognizer configured to: receive input audio comprising speech; transmit at least a portion of the input audio to the network device for processing by the remote speech recognizer to produce a remote speech recognition result; process at least a portion of the input audio to produce a local speech recognition result; identify, based on a command word included in the local speech recognition result, a partial action to be performed; perform, in response to producing the local speech recognition result, the partial action on the electronic device, based, at least in part, on the local speech recognition result; provide an indication of the performed partial action on the electronic device to enable a user of the electronic device to observe the results of the partial action prior to receiving the remote speech recognition result, wherein performing the partial action and providing the indication of the performed partial action are initiated prior to receiving the remote speech recognition result from the network device; and perform a full action on the electronic device that completes the partial action based, at least in part, and responsive to receiving the remote speech recognition result from the network device. | 15. An electronic device for use in a distributed speech recognition system comprising the electronic device and a network device remote from the electronic device, the electronic device, comprising: at least one storage device configured to store one or more applications; an embedded speech recognizer configured to: receive input audio comprising speech; transmit at least a portion of the input audio to the network device for processing by the remote speech recognizer to produce a remote speech recognition result; process at least a portion of the input audio to produce a local speech recognition result; identify, based on a command word included in the local speech recognition result, a partial action to be performed; perform, in response to producing the local speech recognition result, the partial action on the electronic device, based, at least in part, on the local speech recognition result; provide an indication of the performed partial action on the electronic device to enable a user of the electronic device to observe the results of the partial action prior to receiving the remote speech recognition result, wherein performing the partial action and providing the indication of the performed partial action are initiated prior to receiving the remote speech recognition result from the network device; and perform a full action on the electronic device that completes the partial action based, at least in part, and responsive to receiving the remote speech recognition result from the network device. 16. The electronic device of claim 15 , wherein the embedded speech recognizer is configured to: receive from the network device the remote speech recognition result; wherein performing the full action is based, at least in part, on the remote speech recognition result received from the network device. | 0.533275 |
3. The method as recited in claim 1 , wherein the context comprises one or more of: a type of document represented in the optical input; and a content of the document represented in the optical input. | 3. The method as recited in claim 1 , wherein the context comprises one or more of: a type of document represented in the optical input; and a content of the document represented in the optical input. 5. The method as recited in claim 3 , wherein the content is selected from: a driver license number, a vehicle identification number, a phone number, a social security number, a signature, a line item of an invoice, a partial or complete address, a universal resource locator, an insurance group number, a credit card number, a tracking number, a photograph, and a distribution of fields depicted on the document. | 0.718367 |
22. A method performed by one or more processes executing on a computer system, the method comprising: presenting a plurality of data entry fields based on a received file, the received file being based on a markup language, the data entry fields being associated with respective data type identifiers included in the received file, the data type identifiers including respective tags that are based on the markup language and identifying a type of data associated with its corresponding data entry field; receiving user input selecting a data entry field from among the presented plurality of data entry fields; determining a configuration mapping corresponding to the type of data identified by the selected data entry field's data type identifier; reconfiguring a dynamically configurable input device in accordance with the determined configuration mapping, wherein the dynamically configurable input device includes one or more configurable controls, and wherein the key mapping specifies a character corresponding to at least one of the one or more configurable controls; presenting the reconfigured dynamically configurable input device to a user; and receiving input in the selected data entry field from the user via the reconfigured dynamically configurable input device. | 22. A method performed by one or more processes executing on a computer system, the method comprising: presenting a plurality of data entry fields based on a received file, the received file being based on a markup language, the data entry fields being associated with respective data type identifiers included in the received file, the data type identifiers including respective tags that are based on the markup language and identifying a type of data associated with its corresponding data entry field; receiving user input selecting a data entry field from among the presented plurality of data entry fields; determining a configuration mapping corresponding to the type of data identified by the selected data entry field's data type identifier; reconfiguring a dynamically configurable input device in accordance with the determined configuration mapping, wherein the dynamically configurable input device includes one or more configurable controls, and wherein the key mapping specifies a character corresponding to at least one of the one or more configurable controls; presenting the reconfigured dynamically configurable input device to a user; and receiving input in the selected data entry field from the user via the reconfigured dynamically configurable input device. 29. The method of claim 22 , wherein the markup language comprises a Hypertext Markup Language. | 0.644258 |
58. The apparatus claim 57, wherein said dialog manager comprises: a dialog controller constructed and arranged to control said plurality of dialogs in accordance with said selected dialog launch modality, said dialog controller forwarding certain ones of said dialog control requests generated by said graphical user interface to said selected ones of said plurality of dialogs, and to selectively cause said generation of said operating system calls. | 58. The apparatus claim 57, wherein said dialog manager comprises: a dialog controller constructed and arranged to control said plurality of dialogs in accordance with said selected dialog launch modality, said dialog controller forwarding certain ones of said dialog control requests generated by said graphical user interface to said selected ones of said plurality of dialogs, and to selectively cause said generation of said operating system calls. 60. The apparatus claim 58, wherein said dialog manager further comprises: a dialog launch modality determinator constructed and arranged to determine said selected dialog launch modality, wherein said dialog launch modality determinator reconciles a global dialog modality representing a dialog launch modality assigned to the computer-based system and an assigned dialog modality assigned to said selected dialog box in accordance with predetermined selection criteria. | 0.830848 |
1. A method for displaying a selected portion of a data set within a display area of a display device, the data set comprising an element group, the element group including a higher-level element and one or more lower-level elements, the method comprising: processing layout information for the higher-level element to identify that a region occupied by the higher-level element does not correspond with a selected portion of the layout area of the data set; based on the identification, determining not to render the higher-level element for display within the display area; determining whether the higher-level element has a predefined spatial relationship with a first lower-level element of the one or more lower-level elements, the predefined spatial relationship indicating that the region occupied by the first lower-level element falls within the region occupied by the higher-level element of the element group; selecting, on the basis of the determination, whether or not to perform a first process, the first process being selected when it is determined that the higher-level element does not have the predefined spatial relationship with the first lower-level element, wherein: the first process comprises processing layout information for the first lower-level element to identify whether or not the region occupied by the first lower-level element corresponds with the selected portion of the data set, and, when it is identified that the region occupied by the first lower-level element corresponds with the selected portion of the data set, selecting to render the first lower-level element for display in the display area; and when it is determined that the higher-level element has the predefined spatial relationship with the first lower-level element, selecting not to render the first lower-level element for display in the display area. | 1. A method for displaying a selected portion of a data set within a display area of a display device, the data set comprising an element group, the element group including a higher-level element and one or more lower-level elements, the method comprising: processing layout information for the higher-level element to identify that a region occupied by the higher-level element does not correspond with a selected portion of the layout area of the data set; based on the identification, determining not to render the higher-level element for display within the display area; determining whether the higher-level element has a predefined spatial relationship with a first lower-level element of the one or more lower-level elements, the predefined spatial relationship indicating that the region occupied by the first lower-level element falls within the region occupied by the higher-level element of the element group; selecting, on the basis of the determination, whether or not to perform a first process, the first process being selected when it is determined that the higher-level element does not have the predefined spatial relationship with the first lower-level element, wherein: the first process comprises processing layout information for the first lower-level element to identify whether or not the region occupied by the first lower-level element corresponds with the selected portion of the data set, and, when it is identified that the region occupied by the first lower-level element corresponds with the selected portion of the data set, selecting to render the first lower-level element for display in the display area; and when it is determined that the higher-level element has the predefined spatial relationship with the first lower-level element, selecting not to render the first lower-level element for display in the display area. 15. The method according to claim 1 , wherein the layout information includes the size and position of the higher-level element and one or more lower-level elements in the element group within a layout area of the data set. | 0.692198 |
1. An apparatus for adding user-supplied text to a digital still image comprising: (a) a card having a surface and image manipulation instructions printed on said surface in encoded form, at least one of said image manipulation instructions comprising instructions for adding user-supplied text to a digital still image; (b) a text entry device comprising: (i) a user interface adapted to receive text from a user; (ii) a memory adapted to store character set information including character set information defining at least one character set in a non-roman font; and (iii) camera communication means adapted to communicate said user-supplied text and said character set information to a digital still camera; and (c) a digital still camera comprising: (i) a sensor adapted to sense an original digital still image; (ii) a card reader adapted to read said image manipulation instructions stored on said card; (iii) text entry device communication means adapted to receive said user-supplied text and said character set information from said text entry device; (iv) image manipulation means adapted to manipulate said original digital still image in accordance with said image manipulation instructions to form a manipulated digital still image which includes said user-supplied text; and (v) a printer device adapted to print said manipulated digital still image. | 1. An apparatus for adding user-supplied text to a digital still image comprising: (a) a card having a surface and image manipulation instructions printed on said surface in encoded form, at least one of said image manipulation instructions comprising instructions for adding user-supplied text to a digital still image; (b) a text entry device comprising: (i) a user interface adapted to receive text from a user; (ii) a memory adapted to store character set information including character set information defining at least one character set in a non-roman font; and (iii) camera communication means adapted to communicate said user-supplied text and said character set information to a digital still camera; and (c) a digital still camera comprising: (i) a sensor adapted to sense an original digital still image; (ii) a card reader adapted to read said image manipulation instructions stored on said card; (iii) text entry device communication means adapted to receive said user-supplied text and said character set information from said text entry device; (iv) image manipulation means adapted to manipulate said original digital still image in accordance with said image manipulation instructions to form a manipulated digital still image which includes said user-supplied text; and (v) a printer device adapted to print said manipulated digital still image. 2. The apparatus as claimed in claim 1 wherein said card comprises an “Artcard” as described herein. | 0.58644 |
1. A method comprising: receiving, at a computing device, an indication of a gesture, the gesture being entered by a user at a first location of a touchscreen operatively coupled to the computing device; determining, by the computing device, a context of the gesture, based at least in part on the type of the gesture and the type of information that is displayed proximate the first location of the touchscreen when the gesture is entered; responsive to determining, by the computing device, that the gesture is a first type of known gesture, performing, by the computing device and using an application predetermined by the user, an action associated with the context of the gesture; and responsive to determining, by the computing device, that the gesture is a second type of known gesture, outputting, by the computing device, for display, a prompt indicating a plurality of applications for the user to select an application from to perform an action associated with the context of the gesture, wherein the application that is predetermined by the user and each of the plurality of applications indicated by the prompt is a computer program that is executable by the computing device to perform the respective action. | 1. A method comprising: receiving, at a computing device, an indication of a gesture, the gesture being entered by a user at a first location of a touchscreen operatively coupled to the computing device; determining, by the computing device, a context of the gesture, based at least in part on the type of the gesture and the type of information that is displayed proximate the first location of the touchscreen when the gesture is entered; responsive to determining, by the computing device, that the gesture is a first type of known gesture, performing, by the computing device and using an application predetermined by the user, an action associated with the context of the gesture; and responsive to determining, by the computing device, that the gesture is a second type of known gesture, outputting, by the computing device, for display, a prompt indicating a plurality of applications for the user to select an application from to perform an action associated with the context of the gesture, wherein the application that is predetermined by the user and each of the plurality of applications indicated by the prompt is a computer program that is executable by the computing device to perform the respective action. 5. The method of claim 1 , wherein performing the action in response to the first type of known gesture comprises performing, by the computing device, a predetermined system action associated with the context. | 0.640867 |
21. A knowledge based document retrieval system, comprising: a user interface which inputs from a user a series of words, creates a query expression and an internal query condition from a dialogue, including said inputted series of words, resulting from interaction by said user with said system, said query expression and said internal query condition being created based on information related to various concepts, and displays responses from said system and information input by said user; a knowledge base for storing knowledge including said concepts and relations among said concepts, wherein said stored knowledge is represented by concept nodes and relation links forming a concept network, wherein each of said nodes represents a concept and each of said links represents a relation among said concepts; information search means for identifying concept nodes that match said internal query condition semantically; and information retrieval means for retrieving at least one document associated with said identified concept nodes. | 21. A knowledge based document retrieval system, comprising: a user interface which inputs from a user a series of words, creates a query expression and an internal query condition from a dialogue, including said inputted series of words, resulting from interaction by said user with said system, said query expression and said internal query condition being created based on information related to various concepts, and displays responses from said system and information input by said user; a knowledge base for storing knowledge including said concepts and relations among said concepts, wherein said stored knowledge is represented by concept nodes and relation links forming a concept network, wherein each of said nodes represents a concept and each of said links represents a relation among said concepts; information search means for identifying concept nodes that match said internal query condition semantically; and information retrieval means for retrieving at least one document associated with said identified concept nodes. 29. A knowledge based document retrieval system according to claim 21 wherein said user interface implements operations through a multi-window function. | 0.778789 |
15. The non-transitory computer-readable medium of claim 13 , wherein the operations further comprise: identifying, by the computing system, that a second user input selected, from among a third set of search results provided to a computing device responsive to a third query, a particular search result that references the first electronic document; and generating, by the computing system and in response to identifying that the second user input selected the particular search result that references the first electronic document, a second association between the first electronic document and the one or more terms derived from the third query. | 15. The non-transitory computer-readable medium of claim 13 , wherein the operations further comprise: identifying, by the computing system, that a second user input selected, from among a third set of search results provided to a computing device responsive to a third query, a particular search result that references the first electronic document; and generating, by the computing system and in response to identifying that the second user input selected the particular search result that references the first electronic document, a second association between the first electronic document and the one or more terms derived from the third query. 16. The non-transitory computer-readable medium of claim 15 , wherein the operations comprise, after generating the second association between the first electronic document and the one or more terms derived from the third query: receiving a fourth query; determining a relevance of the first electronic document to the fourth query based at least in part on a level of similarity between (i) the one or more terms derived from the third query and associated with the first electronic document and (ii) one or more terms derived from the fourth query; generating a fourth set of search results responsive to the fourth query, including selecting or ranking the first electronic document in the fourth set of search results based on the determined relevance; and transmitting the fourth set of search results. | 0.734237 |
29. A system for managing a collaborative deal closing process that provides a means for tracking and managing signature pages of a closing deal using a taxonomy displayable by a computing device, the system comprising: means for receiving a list of users that are authorized to access the closing deal, the list including an identifier associated with each of the users; means for storing the list of authorized users; means for parsing the identifier associated with each user, wherein the identifier includes one from the group consisting of an email address, a domain name, a name of a company or group, and any combination thereof of the authorized users; means for grouping the users according to parties based on the identifier; means for creating a taxonomy including a listing of documents relevant to the deal and a listing of the parties of the deal; means for receiving at least one document at the server; and means for storing relevant pages of the at least one document, wherein each page is associated with a relevant party in the taxonomy. | 29. A system for managing a collaborative deal closing process that provides a means for tracking and managing signature pages of a closing deal using a taxonomy displayable by a computing device, the system comprising: means for receiving a list of users that are authorized to access the closing deal, the list including an identifier associated with each of the users; means for storing the list of authorized users; means for parsing the identifier associated with each user, wherein the identifier includes one from the group consisting of an email address, a domain name, a name of a company or group, and any combination thereof of the authorized users; means for grouping the users according to parties based on the identifier; means for creating a taxonomy including a listing of documents relevant to the deal and a listing of the parties of the deal; means for receiving at least one document at the server; and means for storing relevant pages of the at least one document, wherein each page is associated with a relevant party in the taxonomy. 52. A system according to claim 29 , further comprising: means for parsing content of the relevant pages; and means for storing the relevant pages in individual files, based on the parsed content. | 0.561613 |
1. A method for generating a Service-Oriented Architecture (SOA) policy based on a context model, comprising: generating, by a computer hardware system, an application scope of the SOA policy; generating, by the computer hardware system, the context model by collecting SOA metadata documents compliant with the application scope of the SOA policy, establishing inter-document references among the SOA metadata documents, and aggregating the SOA metadata documents based on the inter-document references to generate the context model; generating, by the computer hardware system, an action list for the context model based on action semantic modules customized by a user; generating, by the computer hardware system, a condition part of the SOA policy; generating, by the computer hardware system, an action part of the SOA policy according to the action list; and combining, by the computer hardware system, the condition part and the action part of the SOA policy to generate the SOA policy wherein the inter-document references causes the SOA metadata documents to be mutually referenced. | 1. A method for generating a Service-Oriented Architecture (SOA) policy based on a context model, comprising: generating, by a computer hardware system, an application scope of the SOA policy; generating, by the computer hardware system, the context model by collecting SOA metadata documents compliant with the application scope of the SOA policy, establishing inter-document references among the SOA metadata documents, and aggregating the SOA metadata documents based on the inter-document references to generate the context model; generating, by the computer hardware system, an action list for the context model based on action semantic modules customized by a user; generating, by the computer hardware system, a condition part of the SOA policy; generating, by the computer hardware system, an action part of the SOA policy according to the action list; and combining, by the computer hardware system, the condition part and the action part of the SOA policy to generate the SOA policy wherein the inter-document references causes the SOA metadata documents to be mutually referenced. 3. The method of claim 1 , wherein the generating the action list for the context model includes: obtaining the action semantic modules customized by the user and their XML description files; reading the XML description files; and introducing the action semantic modules into the action list according to the XML description files. | 0.592569 |
14. The system of claim 13 , wherein at least two of the plurality of nodes are connected with one another through an edge, wherein navigating the one or more graphs to traverse to the one or more reached nodes connected to the selected node comprises: identifying the one or more edges associated with the selected node in the one or more graphs; and navigating the one or more graphs to reach the one or more reached nodes connected to the selected node through the one or more associated edges. | 14. The system of claim 13 , wherein at least two of the plurality of nodes are connected with one another through an edge, wherein navigating the one or more graphs to traverse to the one or more reached nodes connected to the selected node comprises: identifying the one or more edges associated with the selected node in the one or more graphs; and navigating the one or more graphs to reach the one or more reached nodes connected to the selected node through the one or more associated edges. 15. The system of claim 14 , wherein the one or more graphs are navigated simultaneously through the one or more associated edges to reach the one or more reached nodes connected to the selected node. | 0.960644 |
26. A system as recited in claim 22 , wherein the delivery routine prepares the multimedia content for delivery in a specific delivery format. | 26. A system as recited in claim 22 , wherein the delivery routine prepares the multimedia content for delivery in a specific delivery format. 27. A system as recited in claim 26 , wherein the delivery routine further comprises synchronization routines for synchronizing different multimedia streams, such as the synchronization of audio and video streams. | 0.940536 |
11. A computer system for querying an index of first objects comprised of a plurality of cells index entries and a pool of second objects, comprising: a processor; a storage medium; means for evaluating the index of the first objects to produce a group of one or more candidates based on whether one or more index entries of the first objects satisfy cells designated in a query that respective first objects in the index overlap; means for adding second objects from the pool to said group of candidates to produce an interim group of candidates; means for filtering the interim group of candidates by comparing the query with approximations of the candidates of the interim group with the query to produce filtered candidate objects; means for determining if the filtered candidate objects satisfy the query by comparing the first and second objects corresponding to the filtered candidate objects with the query; and in response to determining that the filtered candidate objects satisfy the query, returning a result. | 11. A computer system for querying an index of first objects comprised of a plurality of cells index entries and a pool of second objects, comprising: a processor; a storage medium; means for evaluating the index of the first objects to produce a group of one or more candidates based on whether one or more index entries of the first objects satisfy cells designated in a query that respective first objects in the index overlap; means for adding second objects from the pool to said group of candidates to produce an interim group of candidates; means for filtering the interim group of candidates by comparing the query with approximations of the candidates of the interim group with the query to produce filtered candidate objects; means for determining if the filtered candidate objects satisfy the query by comparing the first and second objects corresponding to the filtered candidate objects with the query; and in response to determining that the filtered candidate objects satisfy the query, returning a result. 12. The computer system of claim 11 , wherein the index is a grid index comprised of a plurality of grid cells, the first and second objects are geometric shapes, and the second objects are larger than the first objects. | 0.530639 |
1. A method for analyzing one or more scripts contained within a document to determine if the one or more scripts perform one or more predefined functions, the method comprising the steps of: from the one or more scripts contained within the document, identifying one or more relevant scripts that perform the one or more predefined functions; interpreting the one or more relevant scripts; intercepting an external function call from the one or more relevant scripts to a document object model of the document while the one or more relevant scripts are being interpreted, wherein the external function call can only be properly responded to with reference to the document object model; generating, independently of the document object model and prior to a construction of a relevant portion of the document object model that would have been required to properly respond to the intercepted external function call, a generic response to the intercepted external function call that was directed to the document object model; providing the generic response to the intercepted external function call; constructing the document object model after the provision of the generic response only if the provision of the generic response did not enable further operation of the relevant scripts; and providing a specific response, obtained with reference to the constructed document object model, to the external function call if the browser was requested to construct the document object model. | 1. A method for analyzing one or more scripts contained within a document to determine if the one or more scripts perform one or more predefined functions, the method comprising the steps of: from the one or more scripts contained within the document, identifying one or more relevant scripts that perform the one or more predefined functions; interpreting the one or more relevant scripts; intercepting an external function call from the one or more relevant scripts to a document object model of the document while the one or more relevant scripts are being interpreted, wherein the external function call can only be properly responded to with reference to the document object model; generating, independently of the document object model and prior to a construction of a relevant portion of the document object model that would have been required to properly respond to the intercepted external function call, a generic response to the intercepted external function call that was directed to the document object model; providing the generic response to the intercepted external function call; constructing the document object model after the provision of the generic response only if the provision of the generic response did not enable further operation of the relevant scripts; and providing a specific response, obtained with reference to the constructed document object model, to the external function call if the browser was requested to construct the document object model. 2. The method of claim 1 further comprising the steps of: providing a time-centric response, unassociated with a current time, to the external function call if the external function call is a time-centric external function call, the time-centric response causing the one or more relevant scripts to resume operation earlier. | 0.58946 |
10. A computer program product for string pattern conceptualization, particularly for a pattern of words, comprising: a computer readable storage device having computer readable program code embodied therewith, where the computer readable program code when executed on a computer causes the computer to: set a reference string set comprising a plurality of strings; find, inside the reference string set comprising the plurality of strings, at least two specific tuples of substring triples that each comprise a prefix substring, a middle substring, and a suffix substring; and consider each tuple as a candidate for representing a related concept; where: each concatenation of the substring triples is an explicit member of the reference string set; each middle substring of the substring triples is unequal to another middle substring within the substring triples found inside the reference string set; each prefix substring of the substring triples is equal to all other prefix substrings within the substring triples found inside the reference string set; each suffix substring of the substring triples is equal to all other suffix substrings within the substring triples found inside the reference string set; and either the prefix substring of the substring triples or the suffix substring of the substring triples is not empty. | 10. A computer program product for string pattern conceptualization, particularly for a pattern of words, comprising: a computer readable storage device having computer readable program code embodied therewith, where the computer readable program code when executed on a computer causes the computer to: set a reference string set comprising a plurality of strings; find, inside the reference string set comprising the plurality of strings, at least two specific tuples of substring triples that each comprise a prefix substring, a middle substring, and a suffix substring; and consider each tuple as a candidate for representing a related concept; where: each concatenation of the substring triples is an explicit member of the reference string set; each middle substring of the substring triples is unequal to another middle substring within the substring triples found inside the reference string set; each prefix substring of the substring triples is equal to all other prefix substrings within the substring triples found inside the reference string set; each suffix substring of the substring triples is equal to all other suffix substrings within the substring triples found inside the reference string set; and either the prefix substring of the substring triples or the suffix substring of the substring triples is not empty. 11. The computer program product of claim 10 , where the computer readable program code when executed on the computer further causes the computer to rank the candidate tuples according to attributes of at least one of the prefix substring of the substring triples and the suffix substring of the substring triples. | 0.519961 |
7. A client computing device associated with a signer, the client computing device comprising: a processor; and memory coupled to the processor and storing instructions that, when executed by the processor, cause the signer's client computing device to perform operations comprising: receiving an electronic signature document from a client computing device associated with a sender, wherein the electronic signature document is received from the client computing device of the sender by the client computing device of the signer independently of an electronic signature service; installing a code module on the signer's client computing device, the code module received from the electronic signature service and configured to transmit the electronic signature document received by an email client of the signer's client computing device to the electronic signature service in response to a first input received from the signer, transmitting, by the signer's client computing device, the received electronic signature document to the electronic signature service for storage in the electronic signature service; accessing, by the singer's client computing device, the electronic signature document stored in the electronic signature service, wherein accessing the electronic signature document includes signing the stored electronic signature document in response to a second input from the signer; and causing, by the signer's client computing device, the electronic signature service to transmit an email attached with a copy of the stored electronic signature document to the client computing device of the sender using the code module. | 7. A client computing device associated with a signer, the client computing device comprising: a processor; and memory coupled to the processor and storing instructions that, when executed by the processor, cause the signer's client computing device to perform operations comprising: receiving an electronic signature document from a client computing device associated with a sender, wherein the electronic signature document is received from the client computing device of the sender by the client computing device of the signer independently of an electronic signature service; installing a code module on the signer's client computing device, the code module received from the electronic signature service and configured to transmit the electronic signature document received by an email client of the signer's client computing device to the electronic signature service in response to a first input received from the signer, transmitting, by the signer's client computing device, the received electronic signature document to the electronic signature service for storage in the electronic signature service; accessing, by the singer's client computing device, the electronic signature document stored in the electronic signature service, wherein accessing the electronic signature document includes signing the stored electronic signature document in response to a second input from the signer; and causing, by the signer's client computing device, the electronic signature service to transmit an email attached with a copy of the stored electronic signature document to the client computing device of the sender using the code module. 9. The client computing device associated with the signer of claim 7 , wherein the signer's client computing device is a smart phone or a tablet computer. | 0.584687 |
7. A computer implemented method for speech recognition comprising the steps of: (a) initializing parameters of a probabilistic mapping between codes that represent speech sounds and a continuity map; (b) training the parameters of the probabilistic mapping, comprising the steps of: (1) inputting a first set of training speech sounds; (2) representing the first set of training speech sounds as a temporal sequence of the codes; (3) defining a first path through the continuity map for the sequence of codes, where the probablistic mapping defines a conditional probability of the sequence of codes given the first path; (4) finding a smooth path through the continuity map that maximizes the conditional probability of the sequence of codes; (5) repeating steps (b)(1)-(b)(4) over additional training speech sounds: (6) given the smooth paths that represent the sets of training speech sounds, adjusting the probabilistic mapping parameters to increase the conditional probability of the sequences of the codes; (c) inputting a new set of speech sounds; (d) representing the new set of speech sounds by a related sequence of the codes; (e) determining the probability of the sequence of codes representing the new speech sounds given the smooth path that maximizes the path of the code sequences for the training speech sounds; (f) identifying the smooth path having the maximum probability for the sequence of the new set of speech sounds; and (g) outputting the maximum probability value as an indicia of recognition of the sequence of new speech sounds. | 7. A computer implemented method for speech recognition comprising the steps of: (a) initializing parameters of a probabilistic mapping between codes that represent speech sounds and a continuity map; (b) training the parameters of the probabilistic mapping, comprising the steps of: (1) inputting a first set of training speech sounds; (2) representing the first set of training speech sounds as a temporal sequence of the codes; (3) defining a first path through the continuity map for the sequence of codes, where the probablistic mapping defines a conditional probability of the sequence of codes given the first path; (4) finding a smooth path through the continuity map that maximizes the conditional probability of the sequence of codes; (5) repeating steps (b)(1)-(b)(4) over additional training speech sounds: (6) given the smooth paths that represent the sets of training speech sounds, adjusting the probabilistic mapping parameters to increase the conditional probability of the sequences of the codes; (c) inputting a new set of speech sounds; (d) representing the new set of speech sounds by a related sequence of the codes; (e) determining the probability of the sequence of codes representing the new speech sounds given the smooth path that maximizes the path of the code sequences for the training speech sounds; (f) identifying the smooth path having the maximum probability for the sequence of the new set of speech sounds; and (g) outputting the maximum probability value as an indicia of recognition of the sequence of new speech sounds. 8. A method according to claim 7, further including the steps of: collecting the training speech sounds from a known speaker according to a known sequence of words; collecting the new speech sounds from an unknown speaker; and outputting the maximum probability value as an indicia that the unknown speaker is the same as the known speaker. | 0.5 |
9. The computer-implemented method of claim 6 , wherein each graphical element of the at least one graphical element is associated with at least one value of at least one field of at least one type of record stored in the data store. | 9. The computer-implemented method of claim 6 , wherein each graphical element of the at least one graphical element is associated with at least one value of at least one field of at least one type of record stored in the data store. 11. The computer-implemented method of claim 9 , wherein the at least one version of the at least one software project is implemented as at least one stream. | 0.958173 |
1. A computer-implemented method of providing ordered search results within a search engine comprising: retrieving items from a network satisfying search criteria provided by a user and ranking the retrieved items in an initial ranked order based upon the search criteria; determining computer-related activities associated with the user, wherein the computer-related activities comprise social networking activities; ranking the retrieved items based on the computer-related activities, wherein the ranking of the retrieved items is dependent upon whether any retrieved item has been referenced by the social networking activities, and the ranking the retrieved items based on the computer-related activities further includes: determining one or more relevance factors for each retrieved item referenced by the social networking activities, wherein each relevance factor for a retrieved item is based upon a corresponding reference of that retrieved item by the social networking activities; and modifying the initial ranked order of the retrieved items by applying the one or more relevance factors of each corresponding retrieved item to the initial ranked order of that corresponding retrieved item to produce a modified ranking; and providing the search results including the retrieved items in order of the modified ranking. | 1. A computer-implemented method of providing ordered search results within a search engine comprising: retrieving items from a network satisfying search criteria provided by a user and ranking the retrieved items in an initial ranked order based upon the search criteria; determining computer-related activities associated with the user, wherein the computer-related activities comprise social networking activities; ranking the retrieved items based on the computer-related activities, wherein the ranking of the retrieved items is dependent upon whether any retrieved item has been referenced by the social networking activities, and the ranking the retrieved items based on the computer-related activities further includes: determining one or more relevance factors for each retrieved item referenced by the social networking activities, wherein each relevance factor for a retrieved item is based upon a corresponding reference of that retrieved item by the social networking activities; and modifying the initial ranked order of the retrieved items by applying the one or more relevance factors of each corresponding retrieved item to the initial ranked order of that corresponding retrieved item to produce a modified ranking; and providing the search results including the retrieved items in order of the modified ranking. 4. The method of claim 1 , wherein at least one of the retrieved items is from a social networking website. | 0.872038 |
10. The computing device of claim 9 wherein the processing circuit is further configured to determine whether the first field comprising the nested document is of a data type that is associated with the first content type and different than the second content type. | 10. The computing device of claim 9 wherein the processing circuit is further configured to determine whether the first field comprising the nested document is of a data type that is associated with the first content type and different than the second content type. 11. The computing device of claim 10 wherein if the data type of the first field comprising the nested document is associated with the first content type and is different than the second content type, the processing circuit is further configured to determine whether the first parser is able to identify a field and a corresponding data type for the field in the nested document. | 0.899947 |
16. A machine-readable, non-transitory and tangible medium having information recorded thereon for providing translated web content, the information, when read by the machine, causes the machine to perform the following: receiving a request, via a public network connection, from an online user for content in a second language translated from content in a first language; obtaining in response to the request, via a public network connection, the content in the first language from an Internet source that hosts the content in the first language; dividing the obtained content in the first language into a plurality of translatable components; determining, with respect to each of the plurality of translatable components, whether there is a corresponding translated component previously stored; generating the content in the second language by replacing each of a number of translatable components with a corresponding translated component; and sending the content in the second language generated in the generating step to the online user as a response to the request, wherein the online user triggers, via the request, the obtaining, dividing, determining, generating, and sending steps, without participating in any of the steps. | 16. A machine-readable, non-transitory and tangible medium having information recorded thereon for providing translated web content, the information, when read by the machine, causes the machine to perform the following: receiving a request, via a public network connection, from an online user for content in a second language translated from content in a first language; obtaining in response to the request, via a public network connection, the content in the first language from an Internet source that hosts the content in the first language; dividing the obtained content in the first language into a plurality of translatable components; determining, with respect to each of the plurality of translatable components, whether there is a corresponding translated component previously stored; generating the content in the second language by replacing each of a number of translatable components with a corresponding translated component; and sending the content in the second language generated in the generating step to the online user as a response to the request, wherein the online user triggers, via the request, the obtaining, dividing, determining, generating, and sending steps, without participating in any of the steps. 18. The medium of claim 16 , wherein the Internet source is different from a system where the request is received. | 0.612913 |
16. The method as defined in claim 1 , wherein: the solution identifier includes a target that includes a character string that identifies an application used to create the electronic form associated with the document; and discovering the solution comprises discovering the solution using the character string. | 16. The method as defined in claim 1 , wherein: the solution identifier includes a target that includes a character string that identifies an application used to create the electronic form associated with the document; and discovering the solution comprises discovering the solution using the character string. 17. The method as defined in claim 16 , wherein: the discovering the solution comprises discovering the character string in a URL. | 0.973112 |
6. One or more computer-readable media storing computer-executable instructions that, when executed on one or more processors, cause the one or more processors to perform acts comprising: selecting a first set of elements from a group of elements for labeling by one or more human users; receiving a label for each element of the first set of elements from the one or more human users; determining a sample selection bias of the first set of elements; and training a model for predicting labels for each element of the group of elements, the training based at least in part on the labels and on the sample selection bias. | 6. One or more computer-readable media storing computer-executable instructions that, when executed on one or more processors, cause the one or more processors to perform acts comprising: selecting a first set of elements from a group of elements for labeling by one or more human users; receiving a label for each element of the first set of elements from the one or more human users; determining a sample selection bias of the first set of elements; and training a model for predicting labels for each element of the group of elements, the training based at least in part on the labels and on the sample selection bias. 9. One or more computer-readable media as recited in claim 6 , wherein the elements of the group of elements comprise images, documents, audio files, video files, or search queries. | 0.815773 |
10. A non-transitory computer-readable storage medium including instructions that, when executed by a processing unit, cause the processing unit to interpret text segments based on word sense by performing the steps of: parsing a text segment to generate one or more text-based words and related syntactic information; mapping, via a processor, each of the one or more text-based words to at least one concept included in a database and based on a semantic network that includes the at least one concept and one or more relevance ratings associated with the at least one concept, wherein each concept included in the semantic network is associated with a meaning and at least one word; generating a plurality of topics based on the mappings and the syntactic information, wherein each topic includes one or more of the concepts included in the semantic network; for each topic included in the plurality of topics, calculating a topic relevance rating between the topic and at least another topic included in the plurality of topics based on the relevance ratings between the one or more concepts included in the topic and one or more concepts included in the another topic; ranking the plurality of topics based on the topic relevance ratings to generate a ranked topic list; and outputting an element for display based on the ranked topic list. | 10. A non-transitory computer-readable storage medium including instructions that, when executed by a processing unit, cause the processing unit to interpret text segments based on word sense by performing the steps of: parsing a text segment to generate one or more text-based words and related syntactic information; mapping, via a processor, each of the one or more text-based words to at least one concept included in a database and based on a semantic network that includes the at least one concept and one or more relevance ratings associated with the at least one concept, wherein each concept included in the semantic network is associated with a meaning and at least one word; generating a plurality of topics based on the mappings and the syntactic information, wherein each topic includes one or more of the concepts included in the semantic network; for each topic included in the plurality of topics, calculating a topic relevance rating between the topic and at least another topic included in the plurality of topics based on the relevance ratings between the one or more concepts included in the topic and one or more concepts included in the another topic; ranking the plurality of topics based on the topic relevance ratings to generate a ranked topic list; and outputting an element for display based on the ranked topic list. 12. The non-transitory computer-readable storage medium of claim 10 , wherein the semantic network includes a first concept, the first concept is associated with a first phrase, and the first phrase includes a first word and a second word. | 0.633195 |
17. A method comprising: accessing a medical history repository, the medical history repository storing medical history information, the medical history information being associated with a plurality of patients; receiving event audio data, the event audio data being based on verbal utterances, the verbal utterances being associated with a pharmaceutical event, the pharmaceutical event being associated with at least one of the patients; obtaining at least one text string, wherein the at least one text string matches at least one interpretation of the received event audio data, wherein the obtaining the at least one text string includes transforming physical audio data into electrical data via an input device, the obtaining the at least one text string is based on information obtained from a pharmaceutical speech repository, the obtaining the at least one text string is based on information obtained from a speech accent repository, and the obtaining the at least one text string is based on a drug matching function, wherein the at least one text string is associated with a pharmaceutical drug, wherein the drug matching function includes an alternative drug matching function that determines the at least one text string, wherein the determining the at least one text string includes determining a name that is associated with an alternative drug, wherein the alternative drug is associated with the pharmaceutical drug; obtaining medical history information, the obtained medical history information being associated with the at least one of the patients; determining, via at least one device processor, one or more adverse drug event (ADE) alerts, the determining the one or more ADE alerts being based on results of a matching operation that compares a first set of information with ADE attributes, the determining the one or more ADE alerts including receiving the ADE attributes from an ADE repository, wherein the first set of information includes the at least one text string and includes medical history attributes which are associated with the at least one of the patients, the ADE repository being hosted on another system that is configured separately from a hosting system that hosts the medical history repository; and initiating a transmission of an audio alert to an audio output device, wherein the audio alert is associated with the one or more ADE alerts. | 17. A method comprising: accessing a medical history repository, the medical history repository storing medical history information, the medical history information being associated with a plurality of patients; receiving event audio data, the event audio data being based on verbal utterances, the verbal utterances being associated with a pharmaceutical event, the pharmaceutical event being associated with at least one of the patients; obtaining at least one text string, wherein the at least one text string matches at least one interpretation of the received event audio data, wherein the obtaining the at least one text string includes transforming physical audio data into electrical data via an input device, the obtaining the at least one text string is based on information obtained from a pharmaceutical speech repository, the obtaining the at least one text string is based on information obtained from a speech accent repository, and the obtaining the at least one text string is based on a drug matching function, wherein the at least one text string is associated with a pharmaceutical drug, wherein the drug matching function includes an alternative drug matching function that determines the at least one text string, wherein the determining the at least one text string includes determining a name that is associated with an alternative drug, wherein the alternative drug is associated with the pharmaceutical drug; obtaining medical history information, the obtained medical history information being associated with the at least one of the patients; determining, via at least one device processor, one or more adverse drug event (ADE) alerts, the determining the one or more ADE alerts being based on results of a matching operation that compares a first set of information with ADE attributes, the determining the one or more ADE alerts including receiving the ADE attributes from an ADE repository, wherein the first set of information includes the at least one text string and includes medical history attributes which are associated with the at least one of the patients, the ADE repository being hosted on another system that is configured separately from a hosting system that hosts the medical history repository; and initiating a transmission of an audio alert to an audio output device, wherein the audio alert is associated with the one or more ADE alerts. 18. The method of claim 17 , wherein the at least one text string includes one or more of: a name attribute associated with the pharmaceutical drug, a strength attribute associated with the pharmaceutical drug, or a dosage attribute associated with the pharmaceutical drug. | 0.734022 |
9. The method of claim 1 , wherein the objective function comprises a first term and a second term, the first term comprising a square of weighting parameters, and the second term comprising the hyperparameter and a summation, over a set of example pairs, of a loss function. | 9. The method of claim 1 , wherein the objective function comprises a first term and a second term, the first term comprising a square of weighting parameters, and the second term comprising the hyperparameter and a summation, over a set of example pairs, of a loss function. 10. The method of claim 9 , wherein the decision function comprises a first term comprising weighting parameters, and a second term comprising the threshold parameter. | 0.911176 |
16. A method for selecting a next song comprising: extracting, by a computing device, song sequence data from a plurality of published recommendation sources based on at least a subset of the contents of a dictionary; generating, by the computing device, a song graph, the song graph comprising a plurality of song nodes, each song node representing a song and at least a subset of the song nodes sharing an edge with at least one other song node, each such edge representing sequence relationships from the extracted song sequence data; and, selecting, by the computing device, a next song to be presented to a listener by: receiving a first song; selecting as an input node the song node corresponding to the received song; selecting a next song node by performing a random walk from the input node to a song node sharing an edge with the input node; and, selecting as the next song the song associated with the selected next song node. | 16. A method for selecting a next song comprising: extracting, by a computing device, song sequence data from a plurality of published recommendation sources based on at least a subset of the contents of a dictionary; generating, by the computing device, a song graph, the song graph comprising a plurality of song nodes, each song node representing a song and at least a subset of the song nodes sharing an edge with at least one other song node, each such edge representing sequence relationships from the extracted song sequence data; and, selecting, by the computing device, a next song to be presented to a listener by: receiving a first song; selecting as an input node the song node corresponding to the received song; selecting a next song node by performing a random walk from the input node to a song node sharing an edge with the input node; and, selecting as the next song the song associated with the selected next song node. 20. The method of claim 16 further comprising: generating an artist graph, the artist graph comprising a plurality of artist nodes, each artist node representing an artist, at least a subset of the artist nodes comprising metadata and at least a subset of the artist nodes further comprising at least one edge, each edge comprising a weight. | 0.587061 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.