Titles
stringlengths 6
220
| Abstracts
stringlengths 37
3.26k
| Years
int64 1.99k
2.02k
| Categories
stringclasses 1
value |
|---|---|---|---|
Aspect Extraction and Sentiment Classification of Mobile Apps using
App-Store Reviews | Understanding of customer sentiment can be useful for product development. On
top of that if the priorities for the development order can be known, then
development procedure become simpler. This work has tried to address this issue
in the mobile app domain. Along with aspect and opinion extraction this work
has also categorized the extracted aspects ac-cording to their importance. This
can help developers to focus their time and energy at the right place.
| 2,017 | Computation and Language |
Modulating and attending the source image during encoding improves
Multimodal Translation | We propose a new and fully end-to-end approach for multimodal translation
where the source text encoder modulates the entire visual input processing
using conditional batch normalization, in order to compute the most informative
image features for our task. Additionally, we propose a new attention mechanism
derived from this original idea, where the attention model for the visual input
is conditioned on the source text encoder representations. In the paper, we
detail our models as well as the image analysis pipeline. Finally, we report
experimental results. They are, as far as we know, the new state of the art on
three different test sets.
| 2,017 | Computation and Language |
Learning Interpretable Spatial Operations in a Rich 3D Blocks World | In this paper, we study the problem of mapping natural language instructions
to complex spatial actions in a 3D blocks world. We first introduce a new
dataset that pairs complex 3D spatial operations to rich natural language
descriptions that require complex spatial and pragmatic interpretations such as
"mirroring", "twisting", and "balancing". This dataset, built on the simulation
environment of Bisk, Yuret, and Marcu (2016), attains language that is
significantly richer and more complex, while also doubling the size of the
original dataset in the 2D environment with 100 new world configurations and
250,000 tokens. In addition, we propose a new neural architecture that achieves
competitive results while automatically discovering an inventory of
interpretable spatial operations (Figure 5)
| 2,017 | Computation and Language |
Multi-Task Learning for Mental Health using Social Media Text | We introduce initial groundwork for estimating suicide risk and mental health
in a deep learning framework. By modeling multiple conditions, the system
learns to make predictions about suicide risk and mental health at a low false
positive rate. Conditions are modeled as tasks in a multi-task learning (MTL)
framework, with gender prediction as an additional auxiliary task. We
demonstrate the effectiveness of multi-task learning by comparison to a
well-tuned single-task baseline with the same number of parameters. Our best
MTL model predicts potential suicide attempt, as well as the presence of
atypical mental health, with AUC > 0.8. We also find additional large
improvements using multi-task learning on mental health tasks with limited
training data.
| 2,017 | Computation and Language |
Inducing Interpretability in Knowledge Graph Embeddings | We study the problem of inducing interpretability in KG embeddings.
Specifically, we explore the Universal Schema (Riedel et al., 2013) and propose
a method to induce interpretability. There have been many vector space models
proposed for the problem, however, most of these methods don't address the
interpretability (semantics) of individual dimensions. In this work, we study
this problem and propose a method for inducing interpretability in KG
embeddings using entity co-occurrence statistics. The proposed method
significantly improves the interpretability, while maintaining comparable
performance in other KG tasks.
| 2,017 | Computation and Language |
Stochastic Answer Networks for Machine Reading Comprehension | We propose a simple yet robust stochastic answer network (SAN) that simulates
multi-step reasoning in machine reading comprehension. Compared to previous
work such as ReasoNet which used reinforcement learning to determine the number
of steps, the unique feature is the use of a kind of stochastic prediction
dropout on the answer module (final layer) of the neural network during the
training. We show that this simple trick improves robustness and achieves
results competitive to the state-of-the-art on the Stanford Question Answering
Dataset (SQuAD), the Adversarial SQuAD, and the Microsoft MAchine Reading
COmprehension Dataset (MS MARCO).
| 2,018 | Computation and Language |
Contextualized Word Representations for Reading Comprehension | Reading a document and extracting an answer to a question about its content
has attracted substantial attention recently. While most work has focused on
the interaction between the question and the document, in this work we evaluate
the importance of context when the question and document are processed
independently. We take a standard neural architecture for this task, and show
that by providing rich contextualized word representations from a large
pre-trained language model as well as allowing the model to choose between
context-dependent and context-independent word representations, we can obtain
dramatic improvements and reach performance comparable to state-of-the-art on
the competitive SQuAD dataset.
| 2,018 | Computation and Language |
Long-Range Correlation Underlying Childhood Language and Generative
Models | Long-range correlation, a property of time series exhibiting long-term
memory, is mainly studied in the statistical physics domain and has been
reported to exist in natural language. Using a state-of-the-art method for such
analysis, long-range correlation is first shown to occur in long CHILDES data
sets. To understand why, Bayesian generative models of language, originally
proposed in the cognitive scientific domain, are investigated. Among
representative models, the Simon model was found to exhibit surprisingly good
long-range correlation, but not the Pitman-Yor model. Since the Simon model is
known not to correctly reflect the vocabulary growth of natural language, a
simple new model is devised as a conjunct of the Simon and Pitman-Yor models,
such that long-range correlation holds with a correct vocabulary growth rate.
The investigation overall suggests that uniform sampling is one cause of
long-range correlation and could thus have a relation with actual linguistic
processes.
| 2,017 | Computation and Language |
Scale Up Event Extraction Learning via Automatic Training Data
Generation | The task of event extraction has long been investigated in a supervised
learning paradigm, which is bound by the number and the quality of the training
instances. Existing training data must be manually generated through a
combination of expert domain knowledge and extensive human involvement.
However, due to drastic efforts required in annotating text, the resultant
datasets are usually small, which severally affects the quality of the learned
model, making it hard to generalize. Our work develops an automatic approach
for generating training data for event extraction. Our approach allows us to
scale up event extraction training instances from thousands to hundreds of
thousands, and it does this at a much lower cost than a manual approach. We
achieve this by employing distant supervision to automatically create event
annotations from unlabelled text using existing structured knowledge bases or
tables.We then develop a neural network model with post inference to transfer
the knowledge extracted from structured knowledge bases to automatically
annotate typed events with corresponding arguments in text.We evaluate our
approach by using the knowledge extracted from Freebase to label texts from
Wikipedia articles. Experimental results show that our approach can generate a
large number of high quality training instances. We show that this large volume
of training data not only leads to a better event extractor, but also allows us
to detect multiple typed events.
| 2,017 | Computation and Language |
A Novel Way of Identifying Cyber Predators | Recurrent Neural Networks with Long Short-Term Memory cell (LSTM-RNN) have
impressive ability in sequence data processing, particularly for language model
building and text classification. This research proposes the combination of
sentiment analysis, new approach of sentence vectors and LSTM-RNN as a novel
way for Sexual Predator Identification (SPI). LSTM-RNN language model is
applied to generate sentence vectors which are the last hidden states in the
language model. Sentence vectors are fed into another LSTM-RNN classifier, so
as to capture suspicious conversations. Hidden state enables to generate
vectors for sentences never seen before. Fasttext is used to filter the
contents of conversations and generate a sentiment score so as to identify
potential predators. The experiment achieves a record-breaking accuracy and
precision of 100% with recall of 81.10%, exceeding the top-ranked result in the
SPI competition.
| 2,017 | Computation and Language |
On the Benefit of Combining Neural, Statistical and External Features
for Fake News Identification | Identifying the veracity of a news article is an interesting problem while
automating this process can be a challenging task. Detection of a news article
as fake is still an open question as it is contingent on many factors which the
current state-of-the-art models fail to incorporate. In this paper, we explore
a subtask to fake news identification, and that is stance detection. Given a
news article, the task is to determine the relevance of the body and its claim.
We present a novel idea that combines the neural, statistical and external
features to provide an efficient solution to this problem. We compute the
neural embedding from the deep recurrent model, statistical features from the
weighted n-gram bag-of-words model and handcrafted external features with the
help of feature engineering heuristics. Finally, using deep neural layer all
the features are combined, thereby classifying the headline-body news pair as
agree, disagree, discuss, or unrelated. We compare our proposed technique with
the current state-of-the-art models on the fake news challenge dataset. Through
extensive experiments, we find that the proposed model outperforms all the
state-of-the-art techniques including the submissions to the fake news
challenge.
| 2,017 | Computation and Language |
Learning Robust Dialog Policies in Noisy Environments | Modern virtual personal assistants provide a convenient interface for
completing daily tasks via voice commands. An important consideration for these
assistants is the ability to recover from automatic speech recognition (ASR)
and natural language understanding (NLU) errors. In this paper, we focus on
learning robust dialog policies to recover from these errors. To this end, we
develop a user simulator which interacts with the assistant through voice
commands in realistic scenarios with noisy audio, and use it to learn dialog
policies through deep reinforcement learning. We show that dialogs generated by
our simulator are indistinguishable from human generated dialogs, as determined
by human evaluators. Furthermore, preliminary experimental results show that
the learned policies in noisy environments achieve the same execution success
rate with fewer dialog turns compared to fixed rule-based policies.
| 2,017 | Computation and Language |
A Novel Document Generation Process for Topic Detection based on
Hierarchical Latent Tree Models | We propose a novel document generation process based on hierarchical latent
tree models (HLTMs) learned from data. An HLTM has a layer of observed word
variables at the bottom and multiple layers of latent variables on top. For
each document, we first sample values for the latent variables layer by layer
via logic sampling, then draw relative frequencies for the words conditioned on
the values of the latent variables, and finally generate words for the document
using the relative word frequencies. The motivation for the work is to take
word counts into consideration with HLTMs. In comparison with LDA-based
hierarchical document generation processes, the new process achieves
drastically better model fit with much fewer parameters. It also yields more
meaningful topics and topic hierarchies. It is the new state-of-the-art for the
hierarchical topic detection.
| 2,019 | Computation and Language |
Tracing a Loose Wordhood for Chinese Input Method Engine | Chinese input methods are used to convert pinyin sequence or other Latin
encoding systems into Chinese character sentences. For more effective
pinyin-to-character conversion, typical Input Method Engines (IMEs) rely on a
predefined vocabulary that demands manually maintenance on schedule. For the
purpose of removing the inconvenient vocabulary setting, this work focuses on
automatic wordhood acquisition by fully considering that Chinese inputting is a
free human-computer interaction procedure. Instead of strictly defining words,
a loose word likelihood is introduced for measuring how likely a character
sequence can be a user-recognized word with respect to using IME. Then an
online algorithm is proposed to adjust the word likelihood or generate new
words by comparing user true choice for inputting and the algorithm prediction.
The experimental results show that the proposed solution can agilely adapt to
diverse typings and demonstrate performance approaching highly-optimized IME
with fixed vocabulary.
| 2,017 | Computation and Language |
The Zero Resource Speech Challenge 2017 | We describe a new challenge aimed at discovering subword and word units from
raw speech. This challenge is the followup to the Zero Resource Speech
Challenge 2015. It aims at constructing systems that generalize across
languages and adapt to new speakers. The design features and evaluation metrics
of the challenge are presented and the results of seventeen models are
discussed.
| 2,017 | Computation and Language |
Differentiable lower bound for expected BLEU score | In natural language processing tasks performance of the models is often
measured with some non-differentiable metric, such as BLEU score. To use
efficient gradient-based methods for optimization, it is a common workaround to
optimize some surrogate loss function. This approach is effective if
optimization of such loss also results in improving target metric. The
corresponding problem is referred to as loss-evaluation mismatch. In the
present work we propose a method for calculation of differentiable lower bound
of expected BLEU score that does not involve computationally expensive sampling
procedure such as the one required when using REINFORCE rule from reinforcement
learning (RL) framework.
| 2,018 | Computation and Language |
Social Media Writing Style Fingerprint | We present our approach for computer-aided social media text authorship
attribution based on recent advances in short text authorship verification. We
use various natural language techniques to create word-level and
character-level models that act as hidden layers to simulate a simple neural
network. The choice of word-level and character-level models in each layer was
informed through validation performance. The output layer of our system uses an
unweighted majority vote vector to arrive at a conclusion. We also considered
writing bias in social media posts while collecting our training dataset to
increase system robustness. Our system achieved a precision, recall, and
F-measure of 0.82, 0.926 and 0.869 respectively.
| 2,017 | Computation and Language |
Creating New Language and Voice Components for the Updated MaryTTS
Text-to-Speech Synthesis Platform | We present a new workflow to create components for the MaryTTS text-to-speech
synthesis platform, which is popular with researchers and developers, extending
it to support new languages and custom synthetic voices. This workflow replaces
the previous toolkit with an efficient, flexible process that leverages modern
build automation and cloud-hosted infrastructure. Moreover, it is compatible
with the updated MaryTTS architecture, enabling new features and
state-of-the-art paradigms such as synthesis based on deep neural networks
(DNNs). Like MaryTTS itself, the new tools are free, open source software
(FOSS), and promote the use of open data.
| 2,018 | Computation and Language |
A User-Study on Online Adaptation of Neural Machine Translation to Human
Post-Edits | The advantages of neural machine translation (NMT) have been extensively
validated for offline translation of several language pairs for different
domains of spoken and written language. However, research on interactive
learning of NMT by adaptation to human post-edits has so far been confined to
simulation experiments. We present the first user study on online adaptation of
NMT to user post-edits in the domain of patent translation. Our study involves
29 human subjects (translation students) whose post-editing effort and
translation quality were measured on about 4,500 interactions of a human
post-editor and a machine translation system integrating an online adaptive
learning algorithm. Our experimental results show a significant reduction of
human post-editing effort due to online adaptation in NMT according to several
evaluation metrics, including hTER, hBLEU, and KSMR. Furthermore, we found
significant improvements in BLEU/TER between NMT outputs and professional
translations in granted patents, providing further evidence for the advantages
of online adaptive NMT in an interactive setup.
| 2,018 | Computation and Language |
Passing the Brazilian OAB Exam: data preparation and some experiments | In Brazil, all legal professionals must demonstrate their knowledge of the
law and its application by passing the OAB exams, the national bar exams. The
OAB exams therefore provide an excellent benchmark for the performance of legal
information systems since passing the exam would arguably signal that the
system has acquired capacity of legal reasoning comparable to that of a human
lawyer. This article describes the construction of a new data set and some
preliminary experiments on it, treating the problem of finding the
justification for the answers to questions. The results provide a baseline
performance measure against which to evaluate future improvements. We discuss
the reasons to the poor performance and propose next steps.
| 2,017 | Computation and Language |
Rasa: Open Source Language Understanding and Dialogue Management | We introduce a pair of tools, Rasa NLU and Rasa Core, which are open source
python libraries for building conversational software. Their purpose is to make
machine-learning based dialogue management and language understanding
accessible to non-specialist software developers. In terms of design
philosophy, we aim for ease of use, and bootstrapping from minimal (or no)
initial training data. Both packages are extensively documented and ship with a
comprehensive suite of tests. The code is available at
https://github.com/RasaHQ/
| 2,017 | Computation and Language |
Relation Extraction : A Survey | With the advent of the Internet, large amount of digital text is generated
everyday in the form of news articles, research publications, blogs, question
answering forums and social media. It is important to develop techniques for
extracting information automatically from these documents, as lot of important
information is hidden within them. This extracted information can be used to
improve access and management of knowledge hidden in large text corpora.
Several applications such as Question Answering, Information Retrieval would
benefit from this information. Entities like persons and organizations, form
the most basic unit of the information. Occurrences of entities in a sentence
are often linked through well-defined relations; e.g., occurrences of person
and organization in a sentence may be linked through relations such as employed
at. The task of Relation Extraction (RE) is to identify such relations
automatically. In this paper, we survey several important supervised,
semi-supervised and unsupervised RE techniques. We also cover the paradigms of
Open Information Extraction (OIE) and Distant Supervision. Finally, we describe
some of the recent trends in the RE techniques and possible future research
directions. This survey would be useful for three kinds of readers - i)
Newcomers in the field who want to quickly learn about RE; ii) Researchers who
want to know how the various RE techniques evolved over time and what are
possible future research directions and iii) Practitioners who just need to
know which RE technique works best in various settings.
| 2,017 | Computation and Language |
Monotonic Chunkwise Attention | Sequence-to-sequence models with soft attention have been successfully
applied to a wide variety of problems, but their decoding process incurs a
quadratic time and space cost and is inapplicable to real-time sequence
transduction. To address these issues, we propose Monotonic Chunkwise Attention
(MoChA), which adaptively splits the input sequence into small chunks over
which soft attention is computed. We show that models utilizing MoChA can be
trained efficiently with standard backpropagation while allowing online and
linear-time decoding at test time. When applied to online speech recognition,
we obtain state-of-the-art results and match the performance of a model using
an offline soft attention mechanism. In document summarization experiments
where we do not expect monotonic alignments, we show significantly improved
performance compared to a baseline monotonic attention-based model.
| 2,018 | Computation and Language |
Learning to Attend via Word-Aspect Associative Fusion for Aspect-based
Sentiment Analysis | Aspect-based sentiment analysis (ABSA) tries to predict the polarity of a
given document with respect to a given aspect entity. While neural network
architectures have been successful in predicting the overall polarity of
sentences, aspect-specific sentiment analysis still remains as an open problem.
In this paper, we propose a novel method for integrating aspect information
into the neural model. More specifically, we incorporate aspect information
into the neural model by modeling word-aspect relationships. Our novel model,
\textit{Aspect Fusion LSTM} (AF-LSTM) learns to attend based on associative
relationships between sentence words and aspect which allows our model to
adaptively focus on the correct words given an aspect term. This ameliorates
the flaws of other state-of-the-art models that utilize naive concatenations to
model word-aspect similarity. Instead, our model adopts circular convolution
and circular correlation to model the similarity between aspect and words and
elegantly incorporates this within a differentiable neural attention framework.
Finally, our model is end-to-end differentiable and highly related to
convolution-correlation (holographic like) memories. Our proposed neural model
achieves state-of-the-art performance on benchmark datasets, outperforming
ATAE-LSTM by $4\%-5\%$ on average across multiple datasets.
| 2,017 | Computation and Language |
Learning when to skim and when to read | Many recent advances in deep learning for natural language processing have
come at increasing computational cost, but the power of these state-of-the-art
models is not needed for every example in a dataset. We demonstrate two
approaches to reducing unnecessary computation in cases where a fast but weak
baseline classier and a stronger, slower model are both available. Applying an
AUC-based metric to the task of sentiment classification, we find significant
efficiency gains with both a probability-threshold method for reducing
computational cost and one that uses a secondary decision network.
| 2,017 | Computation and Language |
A Novel Approach for Effective Learning in Low Resourced Scenarios | Deep learning based discriminative methods, being the state-of-the-art
machine learning techniques, are ill-suited for learning from lower amounts of
data. In this paper, we propose a novel framework, called simultaneous two
sample learning (s2sL), to effectively learn the class discriminative
characteristics, even from very low amount of data. In s2sL, more than one
sample (here, two samples) are simultaneously considered to both, train and
test the classifier. We demonstrate our approach for speech/music
discrimination and emotion classification through experiments. Further, we also
show the effectiveness of s2sL approach for classification in low-resource
scenario, and for imbalanced data.
| 2,017 | Computation and Language |
Avoiding Echo-Responses in a Retrieval-Based Conversation System | Retrieval-based conversation systems generally tend to highly rank responses
that are semantically similar or even identical to the given conversation
context. While the system's goal is to find the most appropriate response,
rather than the most semantically similar one, this tendency results in
low-quality responses. We refer to this challenge as the echoing problem. To
mitigate this problem, we utilize a hard negative mining approach at the
training stage. The evaluation shows that the resulting model reduces echoing
and achieves better results in terms of Average Precision and Recall@N metrics,
compared to the models trained without the proposed approach.
| 2,018 | Computation and Language |
Sockeye: A Toolkit for Neural Machine Translation | We describe Sockeye (version 1.12), an open-source sequence-to-sequence
toolkit for Neural Machine Translation (NMT). Sockeye is a production-ready
framework for training and applying models as well as an experimental platform
for researchers. Written in Python and built on MXNet, the toolkit offers
scalable training and inference for the three most prominent encoder-decoder
architectures: attentional recurrent neural networks, self-attentional
transformers, and fully convolutional networks. Sockeye also supports a wide
range of optimizers, normalization and regularization techniques, and inference
improvements from current NMT literature. Users can easily run standard
training recipes, explore different model settings, and incorporate new ideas.
In this paper, we highlight Sockeye's features and benchmark it against other
NMT toolkits on two language arcs from the 2017 Conference on Machine
Translation (WMT): English-German and Latvian-English. We report competitive
BLEU scores across all three architectures, including an overall best score for
Sockeye's transformer implementation. To facilitate further comparison, we
release all system outputs and training scripts used in our experiments. The
Sockeye toolkit is free software released under the Apache 2.0 license.
| 2,018 | Computation and Language |
Sentiment Predictability for Stocks | In this work, we present our findings and experiments for stock-market
prediction using various textual sentiment analysis tools, such as mood
analysis and event extraction, as well as prediction models, such as LSTMs and
specific convolutional architectures.
| 2,018 | Computation and Language |
Hierarchical Text Generation and Planning for Strategic Dialogue | End-to-end models for goal-orientated dialogue are challenging to train,
because linguistic and strategic aspects are entangled in latent state vectors.
We introduce an approach to learning representations of messages in dialogues
by maximizing the likelihood of subsequent sentences and actions, which
decouples the semantics of the dialogue utterance from its linguistic
realization. We then use these latent sentence representations for hierarchical
language generation, planning and reinforcement learning. Experiments show that
our approach increases the end-task reward achieved by the model, improves the
effectiveness of long-term planning using rollouts, and allows self-play
reinforcement learning to improve decision making without diverging from human
language. Our hierarchical latent-variable model outperforms previous work both
linguistically and strategically.
| 2,018 | Computation and Language |
Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram
Predictions | This paper describes Tacotron 2, a neural network architecture for speech
synthesis directly from text. The system is composed of a recurrent
sequence-to-sequence feature prediction network that maps character embeddings
to mel-scale spectrograms, followed by a modified WaveNet model acting as a
vocoder to synthesize timedomain waveforms from those spectrograms. Our model
achieves a mean opinion score (MOS) of $4.53$ comparable to a MOS of $4.58$ for
professionally recorded speech. To validate our design choices, we present
ablation studies of key components of our system and evaluate the impact of
using mel spectrograms as the input to WaveNet instead of linguistic, duration,
and $F_0$ features. We further demonstrate that using a compact acoustic
intermediate representation enables significant simplification of the WaveNet
architecture.
| 2,018 | Computation and Language |
NegBio: a high-performance tool for negation and uncertainty detection
in radiology reports | Negative and uncertain medical findings are frequent in radiology reports,
but discriminating them from positive findings remains challenging for
information extraction. Here, we propose a new algorithm, NegBio, to detect
negative and uncertain findings in radiology reports. Unlike previous
rule-based methods, NegBio utilizes patterns on universal dependencies to
identify the scope of triggers that are indicative of negation or uncertainty.
We evaluated NegBio on four datasets, including two public benchmarking corpora
of radiology reports, a new radiology corpus that we annotated for this work,
and a public corpus of general clinical texts. Evaluation on these datasets
demonstrates that NegBio is highly accurate for detecting negative and
uncertain findings and compares favorably to a widely-used state-of-the-art
system NegEx (an average of 9.5% improvement in precision and 5.1% in
F1-score).
| 2,017 | Computation and Language |
Train Once, Test Anywhere: Zero-Shot Learning for Text Classification | Zero-shot Learners are models capable of predicting unseen classes. In this
work, we propose a Zero-shot Learning approach for text categorization. Our
method involves training model on a large corpus of sentences to learn the
relationship between a sentence and embedding of sentence's tags. Learning such
relationship makes the model generalize to unseen sentences, tags, and even new
datasets provided they can be put into same embedding space. The model learns
to predict whether a given sentence is related to a tag or not; unlike other
classifiers that learn to classify the sentence as one of the possible classes.
We propose three different neural networks for the task and report their
accuracy on the test set of the dataset used for training them as well as two
other standard datasets for which no retraining was done. We show that our
models generalize well across new unseen classes in both cases. Although the
models do not achieve the accuracy level of the state of the art supervised
models, yet it evidently is a step forward towards general intelligence in
natural language processing.
| 2,017 | Computation and Language |
Characterizing Political Fake News in Twitter by its Meta-Data | This article presents a preliminary approach towards characterizing political
fake news on Twitter through the analysis of their meta-data. In particular, we
focus on more than 1.5M tweets collected on the day of the election of Donald
Trump as 45th president of the United States of America. We use the meta-data
embedded within those tweets in order to look for differences between tweets
containing fake news and tweets not containing them. Specifically, we perform
our analysis only on tweets that went viral, by studying proxies for users'
exposure to the tweets, by characterizing accounts spreading fake news, and by
looking at their polarization. We found significant differences on the
distribution of followers, the number of URLs on tweets, and the verification
of the users.
| 2,017 | Computation and Language |
Benford's Law and First Letter of Word | A universal First-Letter Law (FLL) is derived and described. It predicts the
percentages of first letters for words in novels. The FLL is akin to Benford's
law (BL) of first digits, which predicts the percentages of first digits in a
data collection of numbers. Both are universal in the sense that FLL only
depends on the numbers of letters in the alphabet, whereas BL only depends on
the number of digits in the base of the number system. The existence of these
types of universal laws appears counter-intuitive. Nonetheless both describe
data very well. Relations to some earlier works are given. FLL predicts that an
English author on the average starts about 16 out of 100 words with the English
letter `t'. This is corroborated by data, yet an author can freely write
anything. Fuller implications and the applicability of FLL remain for the
future.
| 2,018 | Computation and Language |
Deep Learning for Distant Speech Recognition | Deep learning is an emerging technology that is considered one of the most
promising directions for reaching higher levels of artificial intelligence.
Among the other achievements, building computers that understand speech
represents a crucial leap towards intelligent machines. Despite the great
efforts of the past decades, however, a natural and robust human-machine speech
interaction still appears to be out of reach, especially when users interact
with a distant microphone in noisy and reverberant environments. The latter
disturbances severely hamper the intelligibility of a speech signal, making
Distant Speech Recognition (DSR) one of the major open challenges in the field.
This thesis addresses the latter scenario and proposes some novel techniques,
architectures, and algorithms to improve the robustness of distant-talking
acoustic models. We first elaborate on methodologies for realistic data
contamination, with a particular emphasis on DNN training with simulated data.
We then investigate on approaches for better exploiting speech contexts,
proposing some original methodologies for both feed-forward and recurrent
neural networks. Lastly, inspired by the idea that cooperation across different
DNNs could be the key for counteracting the harmful effects of noise and
reverberation, we propose a novel deep learning paradigm called network of deep
neural networks. The analysis of the original concepts were based on extensive
experimental validations conducted on both real and simulated data, considering
different corpora, microphone configurations, environments, noisy conditions,
and ASR tasks.
| 2,017 | Computation and Language |
Query-Based Abstractive Summarization Using Neural Networks | In this paper, we present a model for generating summaries of text documents
with respect to a query. This is known as query-based summarization. We adapt
an existing dataset of news article summaries for the task and train a
pointer-generator model using this dataset. The generated summaries are
evaluated by measuring similarity to reference summaries. Our results show that
a neural network summarization model, similar to existing neural network models
for abstractive summarization, can be constructed to make use of queries to
produce targeted summaries.
| 2,017 | Computation and Language |
Towards a science of human stories: using sentiment analysis and
emotional arcs to understand the building blocks of complex social systems | Given the growing assortment of sentiment measuring instruments, it is
imperative to understand which aspects of sentiment dictionaries contribute to
both their classification accuracy and their ability to provide richer
understanding of texts. Here, we perform detailed, quantitative tests and
qualitative assessments of 6 dictionary-based methods applied, and briefly
examine a further 20 methods. We show that while inappropriate for sentences,
dictionary-based methods are generally robust in their classification accuracy
for longer texts.
Stories often following distinct emotional trajectories, forming patterns
that are meaningful to us. By classifying the emotional arcs for a filtered
subset of 4,803 stories from Project Gutenberg's fiction collection, we find a
set of six core trajectories which form the building blocks of complex
narratives. Of profound scientific interest will be the degree to which we can
eventually understand the full landscape of human stories, and data driven
approaches will play a crucial role.
Finally, we utilize web-scale data from Twitter to study the limits of what
social data can tell us about public health, mental illness, discourse around
the protest movement of #BlackLivesMatter, discourse around climate change, and
hidden networks. We conclude with a review of published works in complex
systems that separately analyze charitable donations, the happiness of words in
10 languages, 100 years of daily temperature data across the United States, and
Australian Rules Football games.
| 2,017 | Computation and Language |
Low Resourced Machine Translation via Morpho-syntactic Modeling: The
Case of Dialectal Arabic | We present the second ever evaluated Arabic dialect-to-dialect machine
translation effort, and the first to leverage external resources beyond a small
parallel corpus. The subject has not previously received serious attention due
to lack of naturally occurring parallel data; yet its importance is evidenced
by dialectal Arabic's wide usage and breadth of inter-dialect variation,
comparable to that of Romance languages. Our results suggest that modeling
morphology and syntax significantly improves dialect-to-dialect translation,
though optimizing such data-sparse models requires consideration of the
linguistic differences between dialects and the nature of available data and
resources. On a single-reference blind test set where untranslated input scores
6.5 BLEU and a model trained only on parallel data reaches 14.6, pivot
techniques and morphosyntactic modeling significantly improve performance to
17.5.
| 2,017 | Computation and Language |
A Chinese Dataset with Negative Full Forms for General Abbreviation
Prediction | Abbreviation is a common phenomenon across languages, especially in Chinese.
In most cases, if an expression can be abbreviated, its abbreviation is used
more often than its fully expanded forms, since people tend to convey
information in a most concise way. For various language processing tasks,
abbreviation is an obstacle to improving the performance, as the textual form
of an abbreviation does not express useful information, unless it's expanded to
the full form. Abbreviation prediction means associating the fully expanded
forms with their abbreviations. However, due to the deficiency in the
abbreviation corpora, such a task is limited in current studies, especially
considering general abbreviation prediction should also include those full form
expressions that do not have valid abbreviations, namely the negative full
forms (NFFs). Corpora incorporating negative full forms for general
abbreviation prediction are few in number. In order to promote the research in
this area, we build a dataset for general Chinese abbreviation prediction,
which needs a few preprocessing steps, and evaluate several different models on
the built dataset. The dataset is available at
https://github.com/lancopku/Chinese-abbreviation-dataset
| 2,017 | Computation and Language |
Detecting Hate Speech in Social Media | In this paper we examine methods to detect hate speech in social media, while
distinguishing this from general profanity. We aim to establish lexical
baselines for this task by applying supervised classification methods using a
recently released dataset annotated for this purpose. As features, our system
uses character n-grams, word n-grams and word skip-grams. We obtain results of
78% accuracy in identifying posts across three classes. Results demonstrate
that the main challenge lies in discriminating profanity and hate speech from
each other. A number of directions for future work are discussed.
| 2,017 | Computation and Language |
word representation or word embedding in Persian text | Text processing is one of the sub-branches of natural language processing.
Recently, the use of machine learning and neural networks methods has been
given greater consideration. For this reason, the representation of words has
become very important. This article is about word representation or converting
words into vectors in Persian text. In this research GloVe, CBOW and skip-gram
methods are updated to produce embedded vectors for Persian words. In order to
train a neural networks, Bijankhan corpus, Hamshahri corpus and UPEC corpus
have been compound and used. Finally, we have 342,362 words that obtained
vectors in all three models for this words. These vectors have many usage for
Persian natural language processing.
| 2,017 | Computation and Language |
HotFlip: White-Box Adversarial Examples for Text Classification | We propose an efficient method to generate white-box adversarial examples to
trick a character-level neural classifier. We find that only a few
manipulations are needed to greatly decrease the accuracy. Our method relies on
an atomic flip operation, which swaps one token for another, based on the
gradients of the one-hot input vectors. Due to efficiency of our method, we can
perform adversarial training which makes the model more robust to attacks at
test time. With the use of a few semantics-preserving constraints, we
demonstrate that HotFlip can be adapted to attack a word-level classifier as
well.
| 2,018 | Computation and Language |
Subword and Crossword Units for CTC Acoustic Models | This paper proposes a novel approach to create an unit set for CTC based
speech recognition systems. By using Byte Pair Encoding we learn an unit set of
an arbitrary size on a given training text. In contrast to using characters or
words as units this allows us to find a good trade-off between the size of our
unit set and the available training data. We evaluate both Crossword units,
that may span multiple word, and Subword units. By combining this approach with
decoding methods using a separate language model we are able to achieve state
of the art results for grapheme based CTC systems.
| 2,018 | Computation and Language |
Analogy Mining for Specific Design Needs | Finding analogical inspirations in distant domains is a powerful way of
solving problems. However, as the number of inspirations that could be matched
and the dimensions on which that matching could occur grow, it becomes
challenging for designers to find inspirations relevant to their needs.
Furthermore, designers are often interested in exploring specific aspects of a
product-- for example, one designer might be interested in improving the
brewing capability of an outdoor coffee maker, while another might wish to
optimize for portability. In this paper we introduce a novel system for
targeting analogical search for specific needs. Specifically, we contribute a
novel analogical search engine for expressing and abstracting specific design
needs that returns more distant yet relevant inspirations than alternate
approaches.
| 2,017 | Computation and Language |
Unsupervised Word Mapping Using Structural Similarities in Monolingual
Embeddings | Most existing methods for automatic bilingual dictionary induction rely on
prior alignments between the source and target languages, such as parallel
corpora or seed dictionaries. For many language pairs, such supervised
alignments are not readily available. We propose an unsupervised approach for
learning a bilingual dictionary for a pair of languages given their
independently-learned monolingual word embeddings. The proposed method exploits
local and global structures in monolingual vector spaces to align them such
that similar words are mapped to each other. We show empirically that the
performance of bilingual correspondents learned using our proposed unsupervised
method is comparable to that of using supervised bilingual correspondents from
a seed dictionary.
| 2,018 | Computation and Language |
DeepNorm-A Deep Learning Approach to Text Normalization | This paper presents an simple yet sophisticated approach to the challenge by
Sproat and Jaitly (2016)- given a large corpus of written text aligned to its
normalized spoken form, train an RNN to learn the correct normalization
function. Text normalization for a token seems very straightforward without
it's context. But given the context of the used token and then normalizing
becomes tricky for some classes. We present a novel approach in which the
prediction of our classification algorithm is used by our sequence to sequence
model to predict the normalized text of the input token. Our approach takes
very less time to learn and perform well unlike what has been reported by
Google (5 days on their GPU cluster). We have achieved an accuracy of 97.62
which is impressive given the resources we use. Our approach is using the best
of both worlds, gradient boosting - state of the art in most classification
tasks and sequence to sequence learning - state of the art in machine
translation. We present our experiments and report results with various
parameter settings.
| 2,017 | Computation and Language |
Any-gram Kernels for Sentence Classification: A Sentiment Analysis Case
Study | Any-gram kernels are a flexible and efficient way to employ bag-of-n-gram
features when learning from textual data. They are also compatible with the use
of word embeddings so that word similarities can be accounted for. While the
original any-gram kernels are implemented on top of tree kernels, we propose a
new approach which is independent of tree kernels and is more efficient. We
also propose a more effective way to make use of word embeddings than the
original any-gram formulation. When applied to the task of sentiment
classification, our new formulation achieves significantly better performance.
| 2,017 | Computation and Language |
The NarrativeQA Reading Comprehension Challenge | Reading comprehension (RC)---in contrast to information retrieval---requires
integrating information and reasoning about events, entities, and their
relations across a full document. Question answering is conventionally used to
assess RC ability, in both artificial agents and children learning to read.
However, existing RC datasets and tasks are dominated by questions that can be
solved by selecting answers using superficial information (e.g., local context
similarity or global term frequency); they thus fail to test for the essential
integrative aspect of RC. To encourage progress on deeper comprehension of
language, we present a new dataset and set of tasks in which the reader must
answer questions about stories by reading entire books or movie scripts. These
tasks are designed so that successfully answering their questions requires
understanding the underlying narrative rather than relying on shallow pattern
matching or salience. We show that although humans solve the tasks easily,
standard RC models struggle on the tasks presented here. We provide an analysis
of the dataset and the challenges it presents.
| 2,017 | Computation and Language |
Improving End-to-End Speech Recognition with Policy Learning | Connectionist temporal classification (CTC) is widely used for maximum
likelihood learning in end-to-end speech recognition models. However, there is
usually a disparity between the negative maximum likelihood and the performance
metric used in speech recognition, e.g., word error rate (WER). This results in
a mismatch between the objective function and metric during training. We show
that the above problem can be mitigated by jointly training with maximum
likelihood and policy gradient. In particular, with policy learning we are able
to directly optimize on the (otherwise non-differentiable) performance metric.
We show that joint training improves relative performance by 4% to 13% for our
end-to-end model as compared to the same model learned through maximum
likelihood. The model achieves 5.53% WER on Wall Street Journal dataset, and
5.42% and 14.70% on Librispeech test-clean and test-other set, respectively.
| 2,017 | Computation and Language |
Improved Regularization Techniques for End-to-End Speech Recognition | Regularization is important for end-to-end speech models, since the models
are highly flexible and easy to overfit. Data augmentation and dropout has been
important for improving end-to-end models in other domains. However, they are
relatively under explored for end-to-end speech models. Therefore, we
investigate the effectiveness of both methods for end-to-end trainable, deep
speech recognition models. We augment audio data through random perturbations
of tempo, pitch, volume, temporal alignment, and adding random noise.We further
investigate the effect of dropout when applied to the inputs of all layers of
the network. We show that the combination of data augmentation and dropout give
a relative performance improvement on both Wall Street Journal (WSJ) and
LibriSpeech dataset of over 20%. Our model performance is also competitive with
other end-to-end speech models on both datasets.
| 2,017 | Computation and Language |
Attentive Memory Networks: Efficient Machine Reading for Conversational
Search | Recent advances in conversational systems have changed the search paradigm.
Traditionally, a user poses a query to a search engine that returns an answer
based on its index, possibly leveraging external knowledge bases and
conditioning the response on earlier interactions in the search session. In a
natural conversation, there is an additional source of information to take into
account: utterances produced earlier in a conversation can also be referred to
and a conversational IR system has to keep track of information conveyed by the
user during the conversation, even if it is implicit.
We argue that the process of building a representation of the conversation
can be framed as a machine reading task, where an automated system is presented
with a number of statements about which it should answer questions. The
questions should be answered solely by referring to the statements provided,
without consulting external knowledge. The time is right for the information
retrieval community to embrace this task, both as a stand-alone task and
integrated in a broader conversational search setting.
In this paper, we focus on machine reading as a stand-alone task and present
the Attentive Memory Network (AMN), an end-to-end trainable machine reading
algorithm. Its key contribution is in efficiency, achieved by having an
hierarchical input encoder, iterating over the input only once. Speed is an
important requirement in the setting of conversational search, as gaps between
conversational turns have a detrimental effect on naturalness. On 20 datasets
commonly used for evaluating machine reading algorithms we show that the AMN
achieves performance comparable to the state-of-the-art models, while using
considerably fewer computations.
| 2,017 | Computation and Language |
A Flexible Approach to Automated RNN Architecture Generation | The process of designing neural architectures requires expert knowledge and
extensive trial and error. While automated architecture search may simplify
these requirements, the recurrent neural network (RNN) architectures generated
by existing methods are limited in both flexibility and components. We propose
a domain-specific language (DSL) for use in automated architecture search which
can produce novel RNNs of arbitrary depth and width. The DSL is flexible enough
to define standard architectures such as the Gated Recurrent Unit and Long
Short Term Memory and allows the introduction of non-standard RNN components
such as trigonometric curves and layer normalization. Using two different
candidate generation techniques, random search with a ranking function and
reinforcement learning, we explore the novel architectures produced by the RNN
DSL for language modeling and machine translation domains. The resulting
architectures do not follow human intuition yet perform well on their targeted
tasks, suggesting the space of usable RNN architectures is far larger than
previously assumed.
| 2,017 | Computation and Language |
Differentially Private Distributed Learning for Language Modeling Tasks | One of the big challenges in machine learning applications is that training
data can be different from the real-world data faced by the algorithm. In
language modeling, users' language (e.g. in private messaging) could change in
a year and be completely different from what we observe in publicly available
data. At the same time, public data can be used for obtaining general knowledge
(i.e. general model of English). We study approaches to distributed fine-tuning
of a general model on user private data with the additional requirements of
maintaining the quality on the general data and minimization of communication
costs. We propose a novel technique that significantly improves prediction
quality on users' language compared to a general model and outperforms gradient
compression methods in terms of communication efficiency. The proposed
procedure is fast and leads to an almost 70% perplexity reduction and 8.7
percentage point improvement in keystroke saving rate on informal English
texts. We also show that the range of tasks our approach is applicable to is
not limited by language modeling only. Finally, we propose an experimental
framework for evaluating differential privacy of distributed training of
language models and show that our approach has good privacy guarantees.
| 2,018 | Computation and Language |
Ethical Questions in NLP Research: The (Mis)-Use of Forensic Linguistics | Ideas from forensic linguistics are now being used frequently in Natural
Language Processing (NLP), using machine learning techniques. While the role of
forensic linguistics was more benign earlier, it is now being used for purposes
which are questionable. Certain methods from forensic linguistics are employed,
without considering their scientific limitations and ethical concerns. While we
take the specific case of forensic linguistics as an example of such trends in
NLP and machine learning, the issue is a larger one and present in many other
scientific and data-driven domains. We suggest that such trends indicate that
some of the applied sciences are exceeding their legal and scientific briefs.
We highlight how carelessly implemented practices are serving to short-circuit
the due processes of law as well breach ethical codes.
| 2,017 | Computation and Language |
An Ensemble Model with Ranking for Social Dialogue | Open-domain social dialogue is one of the long-standing goals of Artificial
Intelligence. This year, the Amazon Alexa Prize challenge was announced for the
first time, where real customers get to rate systems developed by leading
universities worldwide. The aim of the challenge is to converse "coherently and
engagingly with humans on popular topics for 20 minutes". We describe our Alexa
Prize system (called 'Alana') consisting of an ensemble of bots, combining
rule-based and machine learning systems, and using a contextual ranking
mechanism to choose a system response. The ranker was trained on real user
feedback received during the competition, where we address the problem of how
to train on the noisy and sparse feedback obtained during the competition.
| 2,017 | Computation and Language |
Context-aware Path Ranking for Knowledge Base Completion | Knowledge base (KB) completion aims to infer missing facts from existing ones
in a KB. Among various approaches, path ranking (PR) algorithms have received
increasing attention in recent years. PR algorithms enumerate paths between
entity pairs in a KB and use those paths as features to train a model for
missing fact prediction. Due to their good performances and high model
interpretability, several methods have been proposed. However, most existing
methods suffer from scalability (high RAM consumption) and feature explosion
(trains on an exponentially large number of features) problems. This paper
proposes a Context-aware Path Ranking (C-PR) algorithm to solve these problems
by introducing a selective path exploration strategy. C-PR learns global
semantics of entities in the KB using word embedding and leverages the
knowledge of entity semantics to enumerate contextually relevant paths using
bidirectional random walk. Experimental results on three large KBs show that
the path features (fewer in number) discovered by C-PR not only improve
predictive performance but also are more interpretable than existing baselines.
| 2,017 | Computation and Language |
The Character Thinks Ahead: creative writing with deep learning nets and
its stylistic assessment | We discuss how to control outputs from deep learning models of text corpora
so as to create contemporary poetic works. We assess whether these controls are
successful in the immediate sense of creating stylo- metric distinctiveness.
The specific context is our piece The Character Thinks Ahead (2016/17); the
potential applications are broad.
| 2,017 | Computation and Language |
Variational Attention for Sequence-to-Sequence Models | The variational encoder-decoder (VED) encodes source information as a set of
random variables using a neural network, which in turn is decoded into target
data using another neural network. In natural language processing,
sequence-to-sequence (Seq2Seq) models typically serve as encoder-decoder
networks. When combined with a traditional (deterministic) attention mechanism,
the variational latent space may be bypassed by the attention model, and thus
becomes ineffective. In this paper, we propose a variational attention
mechanism for VED, where the attention vector is also modeled as Gaussian
distributed random variables. Results on two experiments show that, without
loss of quality, our proposed method alleviates the bypassing phenomenon as it
increases the diversity of generated sentences.
| 2,018 | Computation and Language |
TFW, DamnGina, Juvie, and Hotsie-Totsie: On the Linguistic and Social
Aspects of Internet Slang | Slang is ubiquitous on the Internet. The emergence of new social contexts
like micro-blogs, question-answering forums, and social networks has enabled
slang and non-standard expressions to abound on the web. Despite this, slang
has been traditionally viewed as a form of non-standard language -- a form of
language that is not the focus of linguistic analysis and has largely been
neglected. In this work, we use UrbanDictionary to conduct the first
large-scale linguistic analysis of slang and its social aspects on the Internet
to yield insights into this variety of language that is increasingly used all
over the world online.
We begin by computationally analyzing the phonological, morphological and
syntactic properties of slang. We then study linguistic patterns in four
specific categories of slang namely alphabetisms, blends, clippings, and
reduplicatives. Our analysis reveals that slang demonstrates extra-grammatical
rules of phonological and morphological formation that markedly distinguish it
from the standard form shedding insight into its generative patterns. Next, we
analyze the social aspects of slang by studying subject restriction and
stereotyping in slang usage. Analyzing tens of thousands of such slang words
reveals that the majority of slang on the Internet belongs to two major
categories: sex and drugs. We also noted that not only is slang usage not
immune to prevalent social biases and prejudices but also reflects such biases
and stereotypes more intensely than the standard variety.
| 2,017 | Computation and Language |
Source-side Prediction for Neural Headline Generation | The encoder-decoder model is widely used in natural language generation
tasks. However, the model sometimes suffers from repeated redundant generation,
misses important phrases, and includes irrelevant entities. Toward solving
these problems we propose a novel source-side token prediction module. Our
method jointly estimates the probability distributions over source and target
vocabularies to capture a correspondence between source and target tokens. The
experiments show that the proposed model outperforms the current
state-of-the-art method in the headline generation task. Additionally, we show
that our method has an ability to learn a reasonable token-wise correspondence
without knowing any true alignments.
| 2,017 | Computation and Language |
Tracking the Diffusion of Named Entities | Existing studies of how information diffuses across social networks have thus
far concentrated on analysing and recovering the spread of deterministic
innovations such as URLs, hashtags, and group membership. However investigating
how mentions of real-world entities appear and spread has yet to be explored,
largely due to the computationally intractable nature of performing large-scale
entity extraction. In this paper we present, to the best of our knowledge, one
of the first pieces of work to closely examine the diffusion of named entities
on social media, using Reddit as our case study platform. We first investigate
how named entities can be accurately recognised and extracted from discussion
posts. We then use these extracted entities to study the patterns of entity
cascades and how the probability of a user adopting an entity (i.e. mentioning
it) is associated with exposures to the entity. We put these pieces together by
presenting a parallelised diffusion model that can forecast the probability of
entity adoption, finding that the influence of adoption between users can be
characterised by their prior interactions -- as opposed to whether the users
propagated entity-adoptions beforehand. Our findings have important
implications for researchers studying influence and language, and for community
analysts who wish to understand entity-level influence dynamics.
| 2,018 | Computation and Language |
Novel Ranking-Based Lexical Similarity Measure for Word Embedding | Distributional semantics models derive word space from linguistic items in
context. Meaning is obtained by defining a distance measure between vectors
corresponding to lexical entities. Such vectors present several problems. In
this paper we provide a guideline for post process improvements to the baseline
vectors. We focus on refining the similarity aspect, address imperfections of
the model by applying the hubness reduction method, implementing relational
knowledge into the model, and providing a new ranking similarity definition
that give maximum weight to the top 1 component value. This feature ranking is
similar to the one used in information retrieval. All these enrichments
outperform current literature results for joint ESL and TOEF sets comparison.
Since single word embedding is a basic element of any semantic task one can
expect a significant improvement of results for these tasks. Moreover, our
improved method of text processing can be translated to continuous distributed
representation of biological sequences for deep proteomics and genomics.
| 2,017 | Computation and Language |
Find the Conversation Killers: a Predictive Study of Thread-ending Posts | How to improve the quality of conversations in online communities has
attracted considerable attention recently. Having engaged, urbane, and reactive
online conversations has a critical effect on the social life of Internet
users. In this study, we are particularly interested in identifying a post in a
multi-party conversation that is unlikely to be further replied to, which
therefore kills that thread of the conversation. For this purpose, we propose a
deep learning model called the ConverNet. ConverNet is attractive due to its
capability of modeling the internal structure of a long conversation and its
appropriate encoding of the contextual information of the conversation, through
effective integration of attention mechanisms. Empirical experiments on
real-world datasets demonstrate the effectiveness of the proposal model. For
the widely concerned topic, our analysis also offers implications for improving
the quality and user experience of online conversations.
| 2,017 | Computation and Language |
Emo, Love, and God: Making Sense of Urban Dictionary, a Crowd-Sourced
Online Dictionary | The Internet facilitates large-scale collaborative projects and the emergence
of Web 2.0 platforms, where producers and consumers of content unify, has
drastically changed the information market. On the one hand, the promise of the
"wisdom of the crowd" has inspired successful projects such as Wikipedia, which
has become the primary source of crowd-based information in many languages. On
the other hand, the decentralized and often un-monitored environment of such
projects may make them susceptible to low quality content. In this work, we
focus on Urban Dictionary, a crowd-sourced online dictionary. We combine
computational methods with qualitative annotation and shed light on the overall
features of Urban Dictionary in terms of growth, coverage and types of content.
We measure a high presence of opinion-focused entries, as opposed to the
meaning-focused entries that we expect from traditional dictionaries.
Furthermore, Urban Dictionary covers many informal, unfamiliar words as well as
proper nouns. Urban Dictionary also contains offensive content, but highly
offensive content tends to receive lower scores through the dictionary's voting
system. The low threshold to include new material in Urban Dictionary enables
quick recording of new words and new meanings, but the resulting heterogeneous
content can pose challenges in using Urban Dictionary as a source to study
language innovation.
| 2,018 | Computation and Language |
Are words easier to learn from infant- than adult-directed speech? A
quantitative corpus-based investigation | We investigate whether infant-directed speech (IDS) could facilitate word
form learning when compared to adult-directed speech (ADS). To study this, we
examine the distribution of word forms at two levels, acoustic and
phonological, using a large database of spontaneous speech in Japanese. At the
acoustic level we show that, as has been documented before for phonemes, the
realizations of words are more variable and less discriminable in IDS than in
ADS. At the phonological level, we find an effect in the opposite direction:
the IDS lexicon contains more distinctive words (such as onomatopoeias) than
the ADS counterpart. Combining the acoustic and phonological metrics together
in a global discriminability score reveals that the bigger separation of
lexical categories in the phonological space does not compensate for the
opposite effect observed at the acoustic level. As a result, IDS word forms are
still globally less discriminable than ADS word forms, even though the effect
is numerically small. We discuss the implication of these findings for the view
that the functional role of IDS is to improve language learnability.
| 2,017 | Computation and Language |
A Framework for Enriching Lexical Semantic Resources with Distributional
Semantics | We present an approach to combining distributional semantic representations
induced from text corpora with manually constructed lexical-semantic networks.
While both kinds of semantic resources are available with high lexical
coverage, our aligned resource combines the domain specificity and availability
of contextual information from distributional models with the conciseness and
high quality of manually crafted lexical networks. We start with a
distributional representation of induced senses of vocabulary terms, which are
accompanied with rich context information given by related lexical items. We
then automatically disambiguate such representations to obtain a full-fledged
proto-conceptualization, i.e. a typed graph of induced word senses. In a final
step, this proto-conceptualization is aligned to a lexical ontology, resulting
in a hybrid aligned resource. Moreover, unmapped induced senses are associated
with a semantic type in order to connect them to the core resource. Manual
evaluations against ground-truth judgments for different stages of our method
as well as an extrinsic evaluation on a knowledge-based Word Sense
Disambiguation benchmark all indicate the high quality of the new hybrid
resource. Additionally, we show the benefits of enriching top-down lexical
knowledge resources with bottom-up distributional information from text for
addressing high-end knowledge acquisition tasks such as cleaning hypernym
graphs and learning taxonomies from scratch.
| 2,017 | Computation and Language |
Dual Long Short-Term Memory Networks for Sub-Character Representation
Learning | Characters have commonly been regarded as the minimal processing unit in
Natural Language Processing (NLP). But many non-latin languages have
hieroglyphic writing systems, involving a big alphabet with thousands or
millions of characters. Each character is composed of even smaller parts, which
are often ignored by the previous work. In this paper, we propose a novel
architecture employing two stacked Long Short-Term Memory Networks (LSTMs) to
learn sub-character level representation and capture deeper level of semantic
meanings. To build a concrete study and substantiate the efficiency of our
neural architecture, we take Chinese Word Segmentation as a research case
example. Among those languages, Chinese is a typical case, for which every
character contains several components called radicals. Our networks employ a
shared radical level embedding to solve both Simplified and Traditional Chinese
Word Segmentation, without extra Traditional to Simplified Chinese conversion,
in such a highly end-to-end way the word segmentation can be significantly
simplified compared to the previous work. Radical level embeddings can also
capture deeper semantic meaning below character level and improve the system
performance of learning. By tying radical and character embeddings together,
the parameter count is reduced whereas semantic knowledge is shared and
transferred between two levels, boosting the performance largely. On 3 out of 4
Bakeoff 2005 datasets, our method surpassed state-of-the-art results by up to
0.4%. Our results are reproducible, source codes and corpora are available on
GitHub.
| 2,018 | Computation and Language |
Building a Sentiment Corpus of Tweets in Brazilian Portuguese | The large amount of data available in social media, forums and websites
motivates researches in several areas of Natural Language Processing, such as
sentiment analysis. The popularity of the area due to its subjective and
semantic characteristics motivates research on novel methods and approaches for
classification. Hence, there is a high demand for datasets on different domains
and different languages. This paper introduces TweetSentBR, a sentiment corpora
for Brazilian Portuguese manually annotated with 15.000 sentences on TV show
domain. The sentences were labeled in three classes (positive, neutral and
negative) by seven annotators, following literature guidelines for ensuring
reliability on the annotation. We also ran baseline experiments on polarity
classification using three machine learning methods, reaching 80.99% on
F-Measure and 82.06% on accuracy in binary classification, and 59.85% F-Measure
and 64.62% on accuracy on three point classification.
| 2,017 | Computation and Language |
Semi-automatic definite description annotation: a first report | Studies in Referring Expression Generation (REG) often make use of corpora of
definite descriptions produced by human subjects in controlled experiments.
Experiments of this kind, which are essential for the study of reference
phenomena and many others, may however include a considerable amount of noise.
Human subjects may easily lack attention, or may simply misunderstand the task
at hand and, as a result, the elicited data may include large proportions of
ambiguous or ill-formed descriptions. In addition to that, REG corpora are
usually collected for the study of semantics-related phenomena, and it is often
the case that the elicited descriptions (and their input contexts) need to be
annotated with their corresponding semantic properties. This, as in many other
fields, may require considerable time and skilled annotators. As a means to
tackle both kinds of difficulties - poor data quality and high annotation costs
- this work discusses a semi-automatic method for the annotation of definite
descriptions produced by human subjects in REG data collection experiments. The
method makes use of simple rules to establish associations between words and
meanings, and is intended to facilitate the design of experiments that produce
REG corpora.
| 2,017 | Computation and Language |
Leveraging Native Language Speech for Accent Identification using Deep
Siamese Networks | The problem of automatic accent identification is important for several
applications like speaker profiling and recognition as well as for improving
speech recognition systems. The accented nature of speech can be primarily
attributed to the influence of the speaker's native language on the given
speech recording. In this paper, we propose a novel accent identification
system whose training exploits speech in native languages along with the
accented speech. Specifically, we develop a deep Siamese network-based model
which learns the association between accented speech recordings and the native
language speech recordings. The Siamese networks are trained with i-vector
features extracted from the speech recordings using either an unsupervised
Gaussian mixture model (GMM) or a supervised deep neural network (DNN) model.
We perform several accent identification experiments using the CSLU Foreign
Accented English (FAE) corpus. In these experiments, our proposed approach
using deep Siamese networks yield significant relative performance improvements
of 15.4 percent on a 10-class accent identification task, over a baseline
DNN-based classification system that uses GMM i-vectors. Furthermore, we
present a detailed error analysis of the proposed accent identification system.
| 2,018 | Computation and Language |
Generative Adversarial Nets for Multiple Text Corpora | Generative adversarial nets (GANs) have been successfully applied to the
artificial generation of image data. In terms of text data, much has been done
on the artificial generation of natural language from a single corpus. We
consider multiple text corpora as the input data, for which there can be two
applications of GANs: (1) the creation of consistent cross-corpus word
embeddings given different word embeddings per corpus; (2) the generation of
robust bag-of-words document embeddings for each corpora. We demonstrate our
GAN models on real-world text data sets from different corpora, and show that
embeddings from both models lead to improvements in supervised learning
problems.
| 2,017 | Computation and Language |
Actionable Email Intent Modeling with Reparametrized RNNs | Emails in the workplace are often intentional calls to action for its
recipients. We propose to annotate these emails for what action its recipient
will take. We argue that our approach of action-based annotation is more
scalable and theory-agnostic than traditional speech-act-based email intent
annotation, while still carrying important semantic and pragmatic information.
We show that our action-based annotation scheme achieves good inter-annotator
agreement. We also show that we can leverage threaded messages from other
domains, which exhibit comparable intents in their conversation, with domain
adaptive RAINBOW (Recurrently AttentIve Neural Bag-Of-Words). On a collection
of datasets consisting of IRC, Reddit, and email, our reparametrized RNNs
outperform common multitask/multidomain approaches on several speech act
related tasks. We also experiment with a minimally supervised scenario of email
recipient action classification, and find the reparametrized RNNs learn a
useful representation.
| 2,017 | Computation and Language |
Mapping to Declarative Knowledge for Word Problem Solving | Math word problems form a natural abstraction to a range of quantitative
reasoning problems, such as understanding financial news, sports results, and
casualties of war. Solving such problems requires the understanding of several
mathematical concepts such as dimensional analysis, subset relationships, etc.
In this paper, we develop declarative rules which govern the translation of
natural language description of these concepts to math expressions. We then
present a framework for incorporating such declarative knowledge into word
problem solving. Our method learns to map arithmetic word problem text to math
expressions, by learning to select the relevant declarative knowledge for each
operation of the solution expression. This provides a way to handle multiple
concepts in the same problem while, at the same time, support interpretability
of the answer expression. Our method models the mapping to declarative
knowledge as a latent variable, thus removing the need for expensive
annotations. Experimental evaluation suggests that our domain knowledge based
solver outperforms all other systems, and that it generalizes better in the
realistic case where the training data it is exposed to is biased in a
different way than the test data.
| 2,017 | Computation and Language |
Advances in Pre-Training Distributed Word Representations | Many Natural Language Processing applications nowadays rely on pre-trained
word representations estimated from large text corpora such as news
collections, Wikipedia and Web Crawl. In this paper, we show how to train
high-quality word vector representations by using a combination of known tricks
that are however rarely used together. The main result of our work is the new
set of publicly available pre-trained models that outperform the current state
of the art by a large margin on a number of tasks.
| 2,017 | Computation and Language |
Letter-Based Speech Recognition with Gated ConvNets | In the recent literature, "end-to-end" speech systems often refer to
letter-based acoustic models trained in a sequence-to-sequence manner, either
via a recurrent model or via a structured output learning approach (such as
CTC). In contrast to traditional phone (or senone)-based approaches, these
"end-to-end'' approaches alleviate the need of word pronunciation modeling, and
do not require a "forced alignment" step at training time. Phone-based
approaches remain however state of the art on classical benchmarks. In this
paper, we propose a letter-based speech recognition system, leveraging a
ConvNet acoustic model. Key ingredients of the ConvNet are Gated Linear Units
and high dropout. The ConvNet is trained to map audio sequences to their
corresponding letter transcriptions, either via a classical CTC approach, or
via a recent variant called ASG. Coupled with a simple decoder at inference
time, our system matches the best existing letter-based systems on WSJ (in word
error rate), and shows near state of the art performance on LibriSpeech.
| 2,019 | Computation and Language |
A Gap-Based Framework for Chinese Word Segmentation via Very Deep
Convolutional Networks | Most previous approaches to Chinese word segmentation can be roughly
classified into character-based and word-based methods. The former regards this
task as a sequence-labeling problem, while the latter directly segments
character sequence into words. However, if we consider segmenting a given
sentence, the most intuitive idea is to predict whether to segment for each gap
between two consecutive characters, which in comparison makes previous
approaches seem too complex. Therefore, in this paper, we propose a gap-based
framework to implement this intuitive idea. Moreover, very deep convolutional
neural networks, namely, ResNets and DenseNets, are exploited in our
experiments. Results show that our approach outperforms the best
character-based and word-based methods on 5 benchmarks, without any further
post-processing module (e.g. Conditional Random Fields) nor beam search.
| 2,017 | Computation and Language |
Improving Text Normalization by Optimizing Nearest Neighbor Matching | Text normalization is an essential task in the processing and analysis of
social media that is dominated with informal writing. It aims to map informal
words to their intended standard forms. Previously proposed text normalization
approaches typically require manual selection of parameters for improved
performance. In this paper, we present an automatic optimizationbased nearest
neighbor matching approach for text normalization. This approach is motivated
by the observation that text normalization is essentially a matching problem
and nearest neighbor matching with an adaptive similarity function is the most
direct procedure for it. Our similarity function incorporates weighted
contributions of contextual, string, and phonetic similarity, and the nearest
neighbor matching involves a minimum similarity threshold. These four
parameters are tuned efficiently using grid search. We evaluate the performance
of our approach on two benchmark datasets. The results demonstrate that
parameter tuning on small sized labeled datasets produce state-of-the-art text
normalization performances. Thus, this approach allows practically easy
construction of evolving domain-specific normalization lexicons
| 2,017 | Computation and Language |
CNN Is All You Need | The Convolution Neural Network (CNN) has demonstrated the unique advantage in
audio, image and text learning; recently it has also challenged Recurrent
Neural Networks (RNNs) with long short-term memory cells (LSTM) in
sequence-to-sequence learning, since the computations involved in CNN are
easily parallelizable whereas those involved in RNN are mostly sequential,
leading to a performance bottleneck. However, unlike RNN, the native CNN lacks
the history sensitivity required for sequence transformation; therefore
enhancing the sequential order awareness, or position-sensitivity, becomes the
key to make CNN the general deep learning model. In this work we introduce an
extended CNN model with strengthen position-sensitivity, called PoseNet. A
notable feature of PoseNet is the asymmetric treatment of position information
in the encoder and the decoder. Experiments shows that PoseNet allows us to
improve the accuracy of CNN based sequence-to-sequence learning significantly,
achieving around 33-36 BLEU scores on the WMT 2014 English-to-German
translation task, and around 44-46 BLEU scores on the English-to-French
translation task.
| 2,017 | Computation and Language |
A Syntactic Approach to Domain-Specific Automatic Question Generation | Factoid questions are questions that require short fact-based answers.
Automatic generation (AQG) of factoid questions from a given text can
contribute to educational activities, interactive question answering systems,
search engines, and other applications. The goal of our research is to generate
factoid source-question-answer triplets based on a specific domain. We propose
a four-component pipeline, which obtains as input a training corpus of
domain-specific documents, along with a set of declarative sentences from the
same domain, and generates as output a set of factoid questions that refer to
the source sentences but are slightly different from them, so that a
question-answering system or a person can be asked a question that requires a
deeper understanding and knowledge than a simple word-matching. Contrary to
existing domain-specific AQG systems that utilize the template-based approach
to question generation, we propose to transform each source sentence into a set
of questions by applying a series of domain-independent rules (a
syntactic-based approach). Our pipeline was evaluated in the domain of cyber
security using a series of experiments on each component of the pipeline
separately and on the end-to-end system. The proposed approach generated a
higher percentage of acceptable questions than a prior state-of-the-art AQG
system.
| 2,017 | Computation and Language |
Toward Continual Learning for Conversational Agents | While end-to-end neural conversation models have led to promising advances in
reducing hand-crafted features and errors induced by the traditional complex
system architecture, they typically require an enormous amount of data due to
the lack of modularity. Previous studies adopted a hybrid approach with
knowledge-based components either to abstract out domain-specific information
or to augment data to cover more diverse patterns. On the contrary, we propose
to directly address the problem using recent developments in the space of
continual learning for neural models. Specifically, we adopt a
domain-independent neural conversational model and introduce a novel neural
continual learning algorithm that allows a conversational agent to accumulate
skills across different tasks in a data-efficient way. To the best of our
knowledge, this is the first work that applies continual learning to
conversation systems. We verified the efficacy of our method through a
conversational skill transfer from either synthetic dialogs or human-human
dialogs to human-computer conversations in a customer support domain.
| 2,018 | Computation and Language |
Corpus specificity in LSA and Word2vec: the role of out-of-domain
documents | Latent Semantic Analysis (LSA) and Word2vec are some of the most widely used
word embeddings. Despite the popularity of these techniques, the precise
mechanisms by which they acquire new semantic relations between words remain
unclear. In the present article we investigate whether LSA and Word2vec
capacity to identify relevant semantic dimensions increases with size of
corpus. One intuitive hypothesis is that the capacity to identify relevant
dimensions should increase as the amount of data increases. However, if corpus
size grow in topics which are not specific to the domain of interest, signal to
noise ratio may weaken. Here we set to examine and distinguish these
alternative hypothesis. To investigate the effect of corpus specificity and
size in word-embeddings we study two ways for progressive elimination of
documents: the elimination of random documents vs. the elimination of documents
unrelated to a specific task. We show that Word2vec can take advantage of all
the documents, obtaining its best performance when it is trained with the whole
corpus. On the contrary, the specialization (removal of out-of-domain
documents) of the training corpus, accompanied by a decrease of dimensionality,
can increase LSA word-representation quality while speeding up the processing
time. Furthermore, we show that the specialization without the decrease in LSA
dimensionality can produce a strong performance reduction in specific tasks.
From a cognitive-modeling point of view, we point out that LSA's word-knowledge
acquisitions may not be efficiently exploiting higher-order co-occurrences and
global relations, whereas Word2vec does.
| 2,018 | Computation and Language |
Disentangled Representations for Manipulation of Sentiment in Text | The ability to change arbitrary aspects of a text while leaving the core
message intact could have a strong impact in fields like marketing and politics
by enabling e.g. automatic optimization of message impact and personalized
language adapted to the receiver's profile. In this paper we take a first step
towards such a system by presenting an algorithm that can manipulate the
sentiment of a text while preserving its semantics using disentangled
representations. Validation is performed by examining trajectories in embedding
space and analyzing transformed sentences for semantic preservation while
expression of desired sentiment shift.
| 2,018 | Computation and Language |
Detecting Cross-Lingual Plagiarism Using Simulated Word Embeddings | Cross-lingual plagiarism (CLP) occurs when texts written in one language are
translated into a different language and used without acknowledging the
original sources. One of the most common methods for detecting CLP requires
online machine translators (such as Google or Microsoft translate) which are
not always available, and given that plagiarism detection typically involves
large document comparison, the amount of translations required would overwhelm
an online machine translator, especially when detecting plagiarism over the
web. In addition, when translated texts are replaced with their synonyms, using
online machine translators to detect CLP would result in poor performance. This
paper addresses the problem of cross-lingual plagiarism detection (CLPD) by
proposing a model that uses simulated word embeddings to reproduce the
predictions of an online machine translator (Google translate) when detecting
CLP. The simulated embeddings comprise of translated words in different
languages mapped in a common space, and replicated to increase the prediction
probability of retrieving the translations of a word (and their synonyms) from
the model. Unlike most existing models, the proposed model does not require
parallel corpora, and accommodates multiple languages (multi-lingual). We
demonstrated the effectiveness of the proposed model in detecting CLP in
standard datasets that contain CLP cases, and evaluated its performance against
a state-of-the-art baseline that relies on online machine translator (T+MA
model). Evaluation results revealed that the proposed model is not only
effective in detecting CLP, it outperformed the baseline. The results indicate
that CLP could be detected with state-of-the-art performances by leveraging the
prediction accuracy of an internet translator with word embeddings, without
relying on internet translators.
| 2,018 | Computation and Language |
Scalable Multi-Domain Dialogue State Tracking | Dialogue state tracking (DST) is a key component of task-oriented dialogue
systems. DST estimates the user's goal at each user turn given the interaction
until then. State of the art approaches for state tracking rely on deep
learning methods, and represent dialogue state as a distribution over all
possible slot values for each slot present in the ontology. Such a
representation is not scalable when the set of possible values are unbounded
(e.g., date, time or location) or dynamic (e.g., movies or usernames).
Furthermore, training of such models requires labeled data, where each user
turn is annotated with the dialogue state, which makes building models for new
domains challenging. In this paper, we present a scalable multi-domain deep
learning based approach for DST. We introduce a novel framework for state
tracking which is independent of the slot value set, and represent the dialogue
state as a distribution over a set of values of interest (candidate set)
derived from the dialogue history or knowledge. Restricting these candidate
sets to be bounded in size addresses the problem of slot-scalability.
Furthermore, by leveraging the slot-independent architecture and transfer
learning, we show that our proposed approach facilitates quick adaptation to
new domains.
| 2,018 | Computation and Language |
Personal Names in Modern Turkey | We analyzed the most common 5000 male and 5000 female Turkish names based on
their etymological, morphological, and semantic attributes. The name statistics
are based on all Turkish citizens who were alive in 2014 and they cover 90% of
all population. To the best of our knowledge, this study is the most
comprehensive data-driven analysis of Turkish personal names. Female names have
a greater diversity than male names (e.g., top 15 male names cover 25% of the
male population, whereas top 28 female names cover 25% of the female
population). Despite their diversity, female names exhibit predictable
patterns. For example, certain roots such as g\"ul and nar (rose and
pomegranate/red, respectively) are used to generate hundreds of unique female
names. Turkish personal names have their origins mainly in Arabic, followed by
Turkish and Persian. We computed overall frequencies of names according to
broad semantic themes that were identified in previous studies. We found that
foreign-origin names such as olga and khaled, pastoral names such as ya\u{g}mur
and deniz (rain and sea, respectively), and names based on fruits and plants
such as filiz and menek\c{s}e (sprout and violet, respectively) are more
frequently observed among females. Among males, names based on animals such as
arslan and yunus (lion and dolphin, respectively) and names based on famous
and/or historical figures such as mustafa kemal and o\u{g}uz ka\u{g}an (founder
of the Turkish Republic and the founder of the Turks in Turkish mythology,
respectively) are observed more frequently.
| 2,018 | Computation and Language |
The CAPIO 2017 Conversational Speech Recognition System | In this paper we show how we have achieved the state-of-the-art performance
on the industry-standard NIST 2000 Hub5 English evaluation set. We explore
densely connected LSTMs, inspired by the densely connected convolutional
networks recently introduced for image classification tasks. We also propose an
acoustic model adaptation scheme that simply averages the parameters of a seed
neural network acoustic model and its adapted version. This method was applied
with the CallHome training corpus and improved individual system performances
by on average 6.1% (relative) against the CallHome portion of the evaluation
set with no performance loss on the Switchboard portion. With RNN-LM rescoring
and lattice combination on the 5 systems trained across three different phone
sets, our 2017 speech recognition system has obtained 5.0% and 9.1% on
Switchboard and CallHome, respectively, both of which are the best word error
rates reported thus far. According to IBM in their latest work to compare human
and machine transcriptions, our reported Switchboard word error rate can be
considered to surpass the human parity (5.1%) of transcribing conversational
telephone speech.
| 2,018 | Computation and Language |
Bidirectional Attention for SQL Generation | Generating structural query language (SQL) queries from natural language is a
long-standing open problem. Answering a natural language question about a
database table requires modeling complex interactions between the columns of
the table and the question. In this paper, we apply the synthesizing approach
to solve this problem. Based on the structure of SQL queries, we break down the
model to three sub-modules and design specific deep neural networks for each of
them. Taking inspiration from the similar machine reading task, we employ the
bidirectional attention mechanisms and character-level embedding with
convolutional neural networks (CNNs) to improve the result. Experimental
evaluations show that our model achieves the state-of-the-art results in
WikiSQL dataset.
| 2,018 | Computation and Language |
Compare, Compress and Propagate: Enhancing Neural Architectures with
Alignment Factorization for Natural Language Inference | This paper presents a new deep learning architecture for Natural Language
Inference (NLI). Firstly, we introduce a new architecture where alignment pairs
are compared, compressed and then propagated to upper layers for enhanced
representation learning. Secondly, we adopt factorization layers for efficient
and expressive compression of alignment vectors into scalar features, which are
then used to augment the base word representations. The design of our approach
is aimed to be conceptually simple, compact and yet powerful. We conduct
experiments on three popular benchmarks, SNLI, MultiNLI and SciTail, achieving
competitive performance on all. A lightweight parameterization of our model
also enjoys a $\approx 3$ times reduction in parameter size compared to the
existing state-of-the-art models, e.g., ESIM and DIIN, while maintaining
competitive performance. Additionally, visual analysis shows that our
propagated features are highly interpretable.
| 2,018 | Computation and Language |
The origins of Zipf's meaning-frequency law | In his pioneering research, G. K. Zipf observed that more frequent words tend
to have more meanings, and showed that the number of meanings of a word grows
as the square root of its frequency. He derived this relationship from two
assumptions: that words follow Zipf's law for word frequencies (a power law
dependency between frequency and rank) and Zipf's law of meaning distribution
(a power law dependency between number of meanings and rank). Here we show that
a single assumption on the joint probability of a word and a meaning suffices
to infer Zipf's meaning-frequency law or relaxed versions. Interestingly, this
assumption can be justified as the outcome of a biased random walk in the
process of mental exploration.
| 2,018 | Computation and Language |
A New Approach for Measuring Sentiment Orientation based on
Multi-Dimensional Vector Space | This study implements a vector space model approach to measure the sentiment
orientations of words. Two representative vectors for positive/negative
polarity are constructed using high-dimensional vec-tor space in both an
unsupervised and a semi-supervised manner. A sentiment ori-entation value per
word is determined by taking the difference between the cosine distances
against the two reference vec-tors. These two conditions (unsupervised and
semi-supervised) are compared against an existing unsupervised method (Turney,
2002). As a result of our experi-ment, we demonstrate that this novel ap-proach
significantly outperforms the pre-vious unsupervised approach and is more
practical and data efficient as well.
| 2,018 | Computation and Language |
Beyond Word Embeddings: Learning Entity and Concept Representations from
Large Scale Knowledge Bases | Text representations using neural word embeddings have proven effective in
many NLP applications. Recent researches adapt the traditional word embedding
models to learn vectors of multiword expressions (concepts/entities). However,
these methods are limited to textual knowledge bases (e.g., Wikipedia). In this
paper, we propose a novel and simple technique for integrating the knowledge
about concepts from two large scale knowledge bases of different structure
(Wikipedia and Probase) in order to learn concept representations. We adapt the
efficient skip-gram model to seamlessly learn from the knowledge in Wikipedia
text and Probase concept graph. We evaluate our concept embedding models on two
tasks: (1) analogical reasoning, where we achieve a state-of-the-art
performance of 91% on semantic analogies, (2) concept categorization, where we
achieve a state-of-the-art performance on two benchmark datasets achieving
categorization accuracy of 100% on one and 98% on the other. Additionally, we
present a case study to evaluate our model on unsupervised argument type
identification for neural semantic parsing. We demonstrate the competitive
accuracy of our unsupervised method and its ability to better generalize to out
of vocabulary entity mentions compared to the tedious and error prone methods
which depend on gazetteers and regular expressions.
| 2,018 | Computation and Language |
PronouncUR: An Urdu Pronunciation Lexicon Generator | State-of-the-art speech recognition systems rely heavily on three basic
components: an acoustic model, a pronunciation lexicon and a language model. To
build these components, a researcher needs linguistic as well as technical
expertise, which is a barrier in low-resource domains. Techniques to construct
these three components without having expert domain knowledge are in great
demand. Urdu, despite having millions of speakers all over the world, is a
low-resource language in terms of standard publically available linguistic
resources. In this paper, we present a grapheme-to-phoneme conversion tool for
Urdu that generates a pronunciation lexicon in a form suitable for use with
speech recognition systems from a list of Urdu words. The tool predicts the
pronunciation of words using a LSTM-based model trained on a handcrafted expert
lexicon of around 39,000 words and shows an accuracy of 64% upon internal
evaluation. For external evaluation on a speech recognition task, we obtain a
word error rate comparable to one achieved using a fully handcrafted expert
lexicon.
| 2,018 | Computation and Language |
Sanskrit Sandhi Splitting using seq2(seq)^2 | In Sanskrit, small words (morphemes) are combined to form compound words
through a process known as Sandhi. Sandhi splitting is the process of splitting
a given compound word into its constituent morphemes. Although rules governing
word splitting exists in the language, it is highly challenging to identify the
location of the splits in a compound word. Though existing Sandhi splitting
systems incorporate these pre-defined splitting rules, they have a low accuracy
as the same compound word might be broken down in multiple ways to provide
syntactically correct splits.
In this research, we propose a novel deep learning architecture called Double
Decoder RNN (DD-RNN), which (i) predicts the location of the split(s) with 95%
accuracy, and (ii) predicts the constituent words (learning the Sandhi
splitting rules) with 79.5% accuracy, outperforming the state-of-art by 20%.
Additionally, we show the generalization capability of our deep learning model,
by showing competitive results in the problem of Chinese word segmentation, as
well.
| 2,019 | Computation and Language |
Automated rating of recorded classroom presentations using speech
analysis in kazakh | Effective presentation skills can help to succeed in business, career and
academy. This paper presents the design of speech assessment during the oral
presentation and the algorithm for speech evaluation based on criteria of
optimal intonation. As the pace of the speech and its optimal intonation varies
from language to language, developing an automatic identification of language
during the presentation is required. Proposed algorithm was tested with
presentations delivered in Kazakh language. For testing purposes the features
of Kazakh phonemes were extracted using MFCC and PLP methods and created a
Hidden Markov Model (HMM) [5], [5] of Kazakh phonemes. Kazakh vowel formants
were defined and the correlation between the deviation rate in fundamental
frequency and the liveliness of the speech to evaluate intonation of the
presentation was analyzed. It was established that the threshold value between
monotone and dynamic speech is 0.16 and the error for intonation evaluation is
19%.
| 2,017 | Computation and Language |
Learning Multimodal Word Representation via Dynamic Fusion Methods | Multimodal models have been proven to outperform text-based models on
learning semantic word representations. Almost all previous multimodal models
typically treat the representations from different modalities equally. However,
it is obvious that information from different modalities contributes
differently to the meaning of words. This motivates us to build a multimodal
model that can dynamically fuse the semantic representations from different
modalities according to different types of words. To that end, we propose three
novel dynamic fusion methods to assign importance weights to each modality, in
which weights are learned under the weak supervision of word association pairs.
The extensive experiments have demonstrated that the proposed methods
outperform strong unimodal baselines and state-of-the-art multimodal models.
| 2,018 | Computation and Language |
Did you hear that? Adversarial Examples Against Automatic Speech
Recognition | Speech is a common and effective way of communication between humans, and
modern consumer devices such as smartphones and home hubs are equipped with
deep learning based accurate automatic speech recognition to enable natural
interaction between humans and machines. Recently, researchers have
demonstrated powerful attacks against machine learning models that can fool
them to produceincorrect results. However, nearly all previous research in
adversarial attacks has focused on image recognition and object detection
models. In this short paper, we present a first of its kind demonstration of
adversarial attacks against speech classification model. Our algorithm performs
targeted attacks with 87% success by adding small background noise without
having to know the underlying model parameter and architecture. Our attack only
changes the least significant bits of a subset of audio clip samples, and the
noise does not change 89% the human listener's perception of the audio clip as
evaluated in our human study.
| 2,018 | Computation and Language |
An Attentive Sequence Model for Adverse Drug Event Extraction from
Biomedical Text | Adverse reaction caused by drugs is a potentially dangerous problem which may
lead to mortality and morbidity in patients. Adverse Drug Event (ADE)
extraction is a significant problem in biomedical research. We model ADE
extraction as a Question-Answering problem and take inspiration from Machine
Reading Comprehension (MRC) literature, to design our model. Our objective in
designing such a model, is to exploit the local linguistic context in clinical
text and enable intra-sequence interaction, in order to jointly learn to
classify drug and disease entities, and to extract adverse reactions caused by
a given drug. Our model makes use of a self-attention mechanism to facilitate
intra-sequence interaction in a text sequence. This enables us to visualize and
understand how the network makes use of the local and wider context for
classification.
| 2,018 | Computation and Language |
Identifying emergency stages in Facebook posts of police departments
with convolutional and recurrent neural networks and support vector machines | Classification of social media posts in emergency response is an important
practical problem: accurate classification can help automate processing of such
messages and help other responders and the public react to emergencies in a
timely fashion. This research focused on classifying Facebook messages of US
police departments. Randomly selected 5,000 messages were used to train
classifiers that distinguished between four categories of messages: emergency
preparedness, response and recovery, as well as general engagement messages.
Features were represented with bag-of-words and word2vec, and models were
constructed using support vector machines (SVMs) and convolutional (CNNs) and
recurrent neural networks (RNNs). The best performing classifier was an RNN
with a custom-trained word2vec model to represent features, which achieved the
F1 measure of 0.839.
| 2,017 | Computation and Language |
Exploring Architectures, Data and Units For Streaming End-to-End Speech
Recognition with RNN-Transducer | We investigate training end-to-end speech recognition models with the
recurrent neural network transducer (RNN-T): a streaming, all-neural,
sequence-to-sequence architecture which jointly learns acoustic and language
model components from transcribed acoustic data. We explore various model
architectures and demonstrate how the model can be improved further if
additional text or pronunciation data are available. The model consists of an
`encoder', which is initialized from a connectionist temporal
classification-based (CTC) acoustic model, and a `decoder' which is partially
initialized from a recurrent neural network language model trained on text data
alone. The entire neural network is trained with the RNN-T loss and directly
outputs the recognized transcript as a sequence of graphemes, thus performing
end-to-end speech recognition. We find that performance can be improved further
through the use of sub-word units (`wordpieces') which capture longer context
and significantly reduce substitution errors. The best RNN-T system, a
twelve-layer LSTM encoder with a two-layer LSTM decoder trained with 30,000
wordpieces as output targets achieves a word error rate of 8.5\% on
voice-search and 5.2\% on voice-dictation tasks and is comparable to a
state-of-the-art baseline at 8.3\% on voice-search and 5.4\% voice-dictation.
| 2,018 | Computation and Language |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.