sentence1
stringlengths 16
446
| sentence2
stringlengths 14
436
|
|---|---|
we prepared pretrained word embeddings using skip-gram model . | we pre-trained word embeddings using word2vec over tweet text of the full training data . |
blitzer et al investigate domain adaptation for pos tagging using the method of structural correspondence learning . | blitzer et al proposed structural correspondence learning to identify the correspondences among features between different domains via the concept of pivot features . |
we use the moses smt toolkit to test the augmented datasets . | we used moses , a phrase-based smt toolkit , for training the translation model . |
coreference resolution is the process of linking together multiple referring expressions of a given entity in the world . | coreference resolution is a multi-faceted task : humans resolve references by exploiting contextual and grammatical clues , as well as semantic information and world knowledge , so capturing each of these will be necessary for an automatic system to fully solve the problem . |
the conll data set was taken from the wall street journal portion of the penn treebank and converted into a dependency format . | the latter was taken from the wall street journal portion of the penn treebank and converted into a dependency format . |
as a measure of the working memory capacity , the japanese version of the reading span test was conducted . | as a measure of the working memory capacity , the japanese version of a reading span test was conducted . |
katz and giesbrecht make use of latent semantic analysis to explore the local linguistic context that can serve to identify multiword expressions that have non-compositional meaning . | katz and giesbrecht use distributional semantics and lsa as a model of context similarity to test whether the local context of a mwe can distinguish its idiomatic use from literal use . |
in this paper , we have studied polarity-bearing topics generated from the jst model and shown that by augmenting the original feature space with polarity-bearing topics , the in-domain supervised classifiers learned from augmented feature representation achieve the state-of-the-art performance . | we study the polarity-bearing topics extracted by jst and show that by augmenting the original feature space with polarity-bearing topics , the in-domain supervised classifiers learned from augmented feature representation achieve the state-of-the-art performance of 95 % on the movie review data and an average of 90 % on the multi-domain sentiment dataset . |
fry , 1955 fry , 1958 showed that intensity was a less effective cue than duration on the perception of linguistic stress patterns . | from a series of experiments , fry , 1955 fry , 1958 showed that duration is a consistent correlate of stress at the word level in english and that it is a more effective cue than intensity . |
subjectivity in natural language refers to aspects of language used to express opinions , feelings , evaluations , and speculations and it , thus , incorporates sentiment . | in natural language , subjectivity refers to expression of opinions , evaluations , feelings , and speculations and thus incorporates sentiment . |
word reordering knowledge needs to be incorporated into attention-based nmt . | we aim to capture word reordering knowledge for the attention-based nmt by incorporating distortion models . |
to address this problem , we proposed the application of the online learning protocol to leverage users feedback and to tailor qe . | to tackle this issue we propose an online framework for adaptive qe that targets reactivity and robustness to user and domain changes . |
transliteration is the task of converting a word from one writing script to another , usually based on the phonetics of the original word . | transliteration is the conversion of a text from one script to another . |
the penn discourse treebank , developed by prasad et al , is currently the largest discourse-annotated corpus , consisting of 2159 wall street journal articles . | the penn discourse tree bank is the largest resource to date that provides a discourse annotated corpus in english . |
hierarchical phrase-based translation was proposed by chiang . | hierarchical phrase-based translation was first proposed by chiang . |
the promt smt system is based on the moses open-source toolkit . | it is a standard phrasebased smt system built using the moses toolkit . |
the systems were tuned using a small extracted parallel dataset with minimum error rate training and then tested with different test sets . | their weights are optimized using minimum error-rate training on a held-out development set for each of the experiments . |
sentiment classification is a useful technique for analyzing subjective information in a large number of texts , and many studies have been conducted ( cite-p-15-3-1 ) . | sentiment classification is a special task of text categorization that aims to classify documents according to their opinion of , or sentiment toward a given subject ( e.g. , if an opinion is supported or not ) ( cite-p-11-1-2 ) . |
we trained a linear log-loss model using stochastic gradient descent learning as implemented in the scikit learn library . | for the feature-based system we used logistic regression classifier from the scikit-learn library . |
but it also eliminates the need to directly predict the direction of translation of the parallel corpus . | an additional advantage of our approach is that it does not require an annotation of the translation direction of the parallel corpus . |
for back-translation , we train a phrase-based smt system for each system in reverse direction . | for preposition and determiner errors , we construct a system using a phrase-based statistical machine translation framework . |
in this work , we use the margin infused relaxed algorithm with a hamming-loss margin . | we select the cutting-plane variant of the margin-infused relaxed algorithm with additional extensions described by eidelman . |
wan et al use a dependency grammar to model word ordering and apply greedy search to find the best permutation . | both wan et al and our system use approximate search to solve the problem of input word ordering . |
word sense disambiguation ( wsd ) is the task of determining the meaning of a word in a given context . | word sense disambiguation ( wsd ) is the nlp task that consists in selecting the correct sense of a polysemous word in a given context . |
the language models were trained using srilm toolkit . | the srilm toolkit was used to build the 5-gram language model . |
distributed representations for words and sentences have been shown to significantly boost the performance of a nlp system . | previous work has shown that unlabeled text can be used to induce unsupervised word clusters which can improve the performance of many supervised nlp tasks . |
semantic parsing is the task of transducing natural language ( nl ) utterances into formal meaning representations ( mrs ) , commonly represented as tree structures . | semantic parsing is the task of converting a sentence into a representation of its meaning , usually in a logical form grounded in the symbols of some fixed ontology or relational database ( cite-p-21-3-3 , cite-p-21-3-4 , cite-p-21-1-11 ) . |
reasoning is the process of thinking in a logical way to form a conclusion . | reasoning is a crucial part of natural language argumentation . |
the most widely used approach works at the word level . | previous such work operates at the word level . |
for building the baseline smt system , we used the open-source smt toolkit moses , in its standard setup . | we compare the final system to moses 3 , an open-source translation toolkit . |
in this study , we proposed a method for disambiguating verbal word senses using term weight learning . | this paper describes unsupervised learning algorithm for disambiguating verbal word senses using term weight learning . |
word sense disambiguation ( wsd ) is a problem of finding the relevant clues in a surrounding context . | word sense disambiguation ( wsd ) is the task of determining the meaning of an ambiguous word in its context . |
n , productions r , start symbol math-w-4-1-0-54 . | truncation size is set to math-w-14-8-0-55 . |
this requires part-of-speech tagging the glosses , for which we use the stanford maximum entropy tagger . | we use stanford log-linear partof-speech tagger to produce pos tags for the english side . |
our algorithm induces a forest of alignments from which we can efficiently extract . | our algorithm yields a forest of word alignments , from which we can efficiently extract the k-best . |
we extract dependency structures from the penn treebank using the head rules of yamada and matsumoto . | we generate dependency structures from the ptb constituency trees using the head rules of yamada and matsumoto . |
finally , we use the bigram similarity dataset from mitchell and lapata which has 3 subsets , adjective-noun , noun-noun , and verbobject , and dev and test sets for each . | specifically , we used the dataset from mitchell and lapata which contains similarity judgments for adjective-noun , noun-noun and verb-object phrases , respectively . |
most relevant to our work is the state of the art in modal sense classification in ruppenhofer and rehbein . | we reconstruct the modal sense classifier of ruppenhofer and rehbein to compare against prior work . |
translation performances are measured with case-insensitive bleu4 score . | translation quality is measured by case-insensitive bleu on newstest13 using one reference translation . |
the knowledge representation system kl-one was the first dl . | the knowledge representation system kl-one , was the first dl . |
for all three classifiers , we used the word2vec 300d pre-trained embeddings as features . | we use the word2vec skip-gram model to train our word embeddings . |
we applied our algorithm to construct a semantic parser for freebase . | to test this capability , we applied the trained parser to natural language queries against freebase . |
we convert the question into a sequence of learned word embeddings by looking up the pre-trained vectors , such as glove . | we initialize the embedding weights by the pre-trained word embeddings with 200 dimensional vectors . |
the models are built using the sri language modeling toolkit . | the language models were trained using srilm toolkit . |
as textual features , we use the pretrained google news word embeddings , obtained by training the skip-gram model with negative sampling . | for our purpose we use word2vec embeddings trained on a google news dataset and find the pairwise cosine distances for all words . |
to evaluate performance we use the second half of the data set released by zeichner , berant , and dagan as a test set . | we use the data set released by zeichner , berant , and dagan , which contains 6,567 entailment rule applications annotated for their validity by crowdsourcing . |
experimental results show that our proposed method outperforms the state-of-the-art methods . | the experimental results show that our method achieves better performance than the state-of-the-art methods . |
we use sri language model toolkit to train a 5-gram model with modified kneser-ney smoothing on the target-side training corpus . | for language model , we use a trigram language model trained with the srilm toolkit on the english side of the training corpus . |
in clark and curran we describe a discriminative method for estimating the parameters of a log-linear parsing model . | in clark and curran we describe efficient methods for performing the calculations using packed charts . |
the key to our solution is the inversion transduction grammars , a type of synchronous context free grammar limiting reordering to adjacent source spans . | most related to our approach , wu used inversion transduction grammars-a synchronous context-free formalism -for this task . |
here we use the discourse relation expansion as defined in the penn discourse treebank . | we use lists of discourse markers compiled from the penn discourse treebank and from to identify such markers in the text . |
estimated on a large set of description-tags pairs , we build a word trigger method ( wtm ) to suggest . | based on this perspective , we build a simple word trigger method ( wtm ) for social tag suggestion . |
our model is based on the standard lstm encoder-decoder model with an attention mechanism . | we use opennmt , which is an implementation of the popular nmt approach that uses an attentional encoder-decoder network . |
the srilm toolkit was used to build this language model . | the language models were trained using srilm toolkit . |
co-training has been successfully applied to various applications , such as statistical parsing and web pages classification . | co-training has been applied to a number of nlp applications , including pos-tagging , parsing , word sense disambiguation , and base noun phrase detection . |
morfessor 2.0 is a new implementation of the morfessor baseline algorithm . | morfessor is a family of probabilistic machine learning methods for finding the morphological segmentation from raw text data . |
in this paper , we have shown the evolution of action recognition datasets and tasks from simple ad-hoc labels . | in this paper , we provide a unified view of action recognition tasks , pointing out their strengths and weaknesses . |
semantic relatedness is a very important factor for coreference resolution , as noun phrases used to refer to the same entity should have a certain semantic relation . | semantic relatedness is the task of quantifying the strength of the semantic connection between textual units , be they words , sentences , or documents . |
erk introduced a distributional similarity-based model for selectional preferences , reminiscent of that of pantel and lin . | in a distributional similarity-based model for selectional preferences is introduced , reminiscent of that of pantel and lin . |
in this paper we develop a baseline approach to identify and verify simple claims about statistical properties . | in this paper we developed a distantly supervised approach for identification and verification of simple statistical claims . |
jeong , lin , and lee use semi-supervised boosting to tag the sentences in e-mail and forum discussions with speech acts by inducing knowledge from annotated spoken conversations . | jeong et al use semi-supervised learning to transfer dialogue acts from labeled speech corpora to the internet media of forums and e-mail . |
we utilize minimum error rate training to optimize feature weights of the paraphrasing model according to ndcg . | we tune phrase-based smt models using minimum error rate training and the development data for each language pair . |
in this paper , we study the problem of obtaining partial annotation from freely available data . | in this paper , we investigate techniques for adopting freely available data to help improve the performance on chinese word segmentation . |
neural networks have been successfully applied to nlp problems , specifically , sequence-to-sequence or models applied to machine translation and word-to-vector . | recurrent neural networks have successfully been used in sequence learning problems , for example machine translation , and language modeling . |
we train probabilistic parsing models for resource-poor languages by transferring cross-lingual knowledge from resource-rich language . | we train probabilistic parsing models for resource-poor languages by maximizing a combination of likelihood on parallel data and confidence on unlabeled data . |
a 5-gram language model of the target language was trained using kenlm . | an english 5-gram language model is trained using kenlm on the gigaword corpus . |
the paper presents an application of structural correspondence learning ( scl ) ( cite-p-14-1-4 ) . | the paper presents an application of structural correspondence learning ( scl ) to parse disambiguation . |
we trained a 5-gram language model on the english side of each training corpus using the sri language modeling toolkit . | we trained a 4-gram language model on the xinhua portion of gigaword corpus using the sri language modeling toolkit with modified kneser-ney smoothing . |
to tackle this problem , hochreiter et al introduced an architecture , called long short-term memory that allows to preserve temporal information , even if the correlated events are separated by a longer time . | to tackle this problem , hochreiter and schmidhuber proposed long short term memory , which uses a cell with input , forget and output gates to prevent the vanishing gradient problem . |
classifier we use the l2-regularized logistic regression from the liblinear package , which we accessed through weka . | we use the multi-class logistic regression classifier from the liblinear package 2 for the prediction of edit scripts . |
relation extraction is the task of finding semantic relations between entities from text . | relation extraction is a fundamental task in information extraction . |
the log-linear feature weights are tuned with minimum error rate training on bleu . | feature weights are tuned using minimum error rate training on the 455 provided references . |
headden , johnson and mcclosky introduced the extended valence grammar and added lexicalization and smoothing . | headden iii et al introduce the extended valence grammar and add lexicalization and smoothing . |
chelba and acero use the parameters of the source domain maximum entropy classifier as the means of a gaussian prior when training a new model on the target data . | chelba and acero use the parameters of the maximum entropy model learned from the source domain as the means of a gaussian prior when training a new model on the target data . |
table 4 shows the comparison of the performances on bleu metric . | the table also shows the popular bleu and nist 2 mt metrics . |
then we train word2vec to represent each entity with a 100-dimensional embedding vector . | we then used word2vec to train word embeddings with 512 dimensions on each of the prepared corpora . |
by casting pseudo-word searching problem into a parsing framework , we search for pseudowords . | by casting pseudo-word searching problem into a parsing framework , we search for pseudowords in polynomial time . |
in contrast to previous statistical learning approaches , we directly translate math word problems . | in contrast to these approaches , we study the feasibility of applying deep learning to the task of math word problem solving . |
in this paper , we propose a method to jointly model and exploit the context compatibility , the topic . | in this paper , we propose a generative model ¨c called entity-topic model , to effectively join the above two complementary directions together . |
the vectors are given by a word2vec model and a glove model trained on german data . | the vectors can be pretrained by neural language models . |
in , kwon et al drew a two-dimensional plot of 59 features ranked by means of forward selection and backward elimination . | in , kwon et al drawled a two-dimensional plot of 59 features ranked by forward selection and backward elimination . |
for the language model , we used srilm with modified kneser-ney smoothing . | we build an open-vocabulary language model with kneser-ney smoothing using the srilm toolkit . |
we use srilm train a 5-gram language model on the xinhua portion of the english gigaword corpus 5th edition with modified kneser-ney discounting . | we train a 4-gram language model on the xinhua portion of the english gigaword corpus using the srilm toolkits with modified kneser-ney smoothing . |
relation extraction ( re ) is the task of recognizing the assertion of a particular relationship between two or more entities in text . | relation extraction ( re ) is the task of assigning a semantic relationship between a pair of arguments . |
parameters do produce useful models of student learning . | user affect parameters can increase the usefulness of these models . |
in this study , we focus on the problem of cross-lingual sentiment classification , which leverages only english training data for supervised sentiment classification of chinese product reviews . | in this study , we focus on improving the corpus-based method for cross-lingual sentiment classification of chinese product reviews by developing novel approaches . |
the expectationmaximization algorithm can be used to train probabilities if the state behaviour is fixed . | for unsupervised learning one can consider the labels as missing data and estimate their values using the expectation maximization algorithm . |
we used word2vec to convert each word in the world state , query to its vector representation . | we initialize our word vectors with 300-dimensional word2vec word embeddings . |
plagiarism is a very significant problem nowadays , specifically in higher education institutions . | plagiarism is a major issue in science and education . |
sentiment analysis is a technique to classify documents based on the polarity of opinion expressed by the author of the document ( cite-p-16-1-13 ) . | sentiment analysis is the computational analysis of people ’ s feelings or beliefs expressed in texts such as emotions , opinions , attitudes , appraisals , etc . ( cite-p-11-3-3 ) . |
we used a 5-gram language model with modified kneser-ney smoothing , built with the srilm toolkit . | we apply sri language modeling toolkit to train a 4-gram language model with kneser-ney smoothing . |
we used a 5-gram language model with modified kneser-ney smoothing , built with the srilm toolkit . | we used kenlm with srilm to train a 5-gram language model based on all available target language training data . |
coreference resolution is a set partitioning problem in which each resulting partition refers to an entity . | coreference resolution is a well known clustering task in natural language processing . |
dependency parsing is a crucial component of many natural language processing systems , for tasks such as text classification ( o ? zgu ? r and gu ? ngo ? r , 2010 ) , statistical machine translation ( cite-p-13-3-0 ) , relation extraction ( cite-p-13-1-1 ) , and question answering ( cite-p-13-1-3 ) . | dependency parsing is the task to assign dependency structures to a given sentence math-w-4-1-0-14 . |
the similarity-based model showed error rates down to 0 . 16 , far lower than both em-based clustering and resnik ’ s wordnet model . | in the evaluation , the similarity-model shows lower error rates than both resnik ’ s wordnet-based model and the em-based clustering model . |
the target-side language models were estimated using the srilm toolkit . | the language models were trained using srilm toolkit . |
semantic role labeling ( srl ) is a kind of shallow semantic parsing task and its goal is to recognize some related phrases and assign a joint structure ( who did what to whom , when , where , why , how ) to each predicate of a sentence ( cite-p-24-3-4 ) . | semantic role labeling ( srl ) is a task of automatically identifying semantic relations between predicate and its related arguments in the sentence . |
we apply the stanford coreference resolution system . | we use the stanford rule-based system for coreference resolution . |
neural models , with various neural architectures , have recently achieved great success . | recently , neural networks become popular for natural language processing . |
with regard to surface realisation , decisions are often made according to a language model of the domain . | surface realisation decisions in a natural language generation system are often made according to a language model of the domain . |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.