text stringlengths 82 736 | label int64 0 1 |
|---|---|
we compare the proposed method particularly with kudo and matsumoto with the same feature set---in this paper , we use the features proposed in kudo and matsumoto | 1 |
likewise , we propose to encode multiple hypotheses into a confusion forest , which is a packed forest which represents multiple parse trees in a polynomial space ments among parse trees---instead of constructing a string-based confusion network , we generate a packed forest which encodes exponentially many parse trees in a polynomial space | 1 |
word2vec , is a neural network model which implements a language model objective---word2vec is the method to obtain distributed representations for a word by using neural networks with one hidden layer | 1 |
we used a standard pbmt system built using moses toolkit---we used moses as the implementation of the baseline smt systems | 1 |
semantic parsing is the task of mapping natural language to machine interpretable meaning representations---we use 300-dimensional word embeddings from glove to initialize the model | 0 |
in addition , a 5-gram lm with kneser-ney smoothing and interpolation was built using the srilm toolkit---the n-gram models are created using the srilm toolkit with good-turning smoothing for both the chinese and english data | 1 |
the reader is referred to for a detailed description of the acoustic analysis procedure---the final smt system performance is evaluated on a uncased test set of 3071 sentences using the bleu , nist and meteor | 0 |
in this paper , we present gated self-matching networks for reading comprehension and question answering---in this paper , we focus on reading comprehension style question answering which aims to answer questions | 1 |
frame induction is the automatic creation of frame-semantic resources similar to framenet or propbank , which map lexical units of a language to frame representations of each lexical unit ’ s semantics---the process of creating a frame lexicon automatically is known as frame induction | 1 |
we use moses , a statistical machine translation system that allows training of translation models---davidov and rappoport proposed a method that detects function words by their high frequency , and utilizes these words for the discovery of symmetric patterns | 0 |
in addition , our enhancements to u-compare mean that various types of multilingual and multimodal workflows can now be created with the minimum effort---coverage and speed , this paper proposes a new web parallel data mining scheme | 0 |
garfield , 1965 ) is probably the first to discuss an automatic computation of citation types---garfield is probably the first to discuss an automatic computation of a citation classification | 1 |
since our dataset is not so large , we make use of pre-trained word embeddings , which are trained on a much larger corpus with word2vec toolkit---thus , we pre-train the embeddings on a huge unlabeled data , the chinese wikipedia corpus , with word2vec toolkit | 1 |
chambers and jurafsky learn narrative schemas , which mean coherent sequences or sets of events , from unlabeled corpora---for instance , chambers and jurafsky model narrative flow in the style of schankian scripts | 1 |
word sense disambiguation ( wsd ) is the task to identify the intended sense of a word in a computational manner based on the context in which it appears ( cite-p-13-3-4 )---word sense disambiguation ( wsd ) is a key task in computational lexical semantics , inasmuch as it addresses the lexical ambiguity of text by making explicit the meaning of words occurring in a given context ( cite-p-18-3-10 ) | 1 |
we use 4-gram language models in both tasks , and conduct minimumerror-rate training to optimize feature weights on the dev set---for language model , we train a 5-gram modified kneser-ney language model and use minimum error rate training to tune the smt | 1 |
a confusion network consists of a sequence of sets of candidate words---a confusion network comprises a sequence of sets of alternative words , possibly including null ’ s , with associated scores | 1 |
this work lays the foundation for automated assessments of narrative quality in student writing---work makes a first attempt at investigating the evaluation of narrative quality | 1 |
first , we use the standard rst-dt corpus that contains discourse annotations for 385 wall street journal news articles from the penn treebank---we compute these using the manual parse annotations for the articles from the penn treebank corpus | 1 |
this paper presents an efficient k-best parsing algorithm for pcfgs---this paper presents an alternative way of pruning unnecessary edges | 1 |
coreference resolution is the process of linking together multiple referring expressions of a given entity in the world---coreference resolution is the task of grouping mentions to entities | 1 |
we employ dropout to mitigate overfitting , and early-stopping---we employ dropout and early-stopping to mitigate overfitting | 1 |
sagan is a semantic textual similarity metric based on a complex textual entailment pipeline---as word vectors the authors use word2vec embeddings trained with the skip-gram model | 0 |
we also demonstrate applicability to other languages and domains---in our experiments , we also demonstrate the applicability of our approach to another language | 1 |
each of the systems is tuned on the development set , and blind evaluation is performed on the test set---the system is tuned on the development data , and finally blind evaluation is performed on the test data | 1 |
the feature weights for each system were tuned on development sets using the moses implementation of minimum error rate training---we trained the statistical phrase-based systems using the moses toolkit with mert tuning | 0 |
we first train a word2vec model on fr-wikipedia 11 to obtain non contextual word vectors---text summarization is the process of creating a compressed version of a given document that delivers the main topic of the document | 0 |
furthermore , we train a 5-gram language model using the sri language toolkit---we trained a 5-grams language model by the srilm toolkit | 1 |
relation extraction ( re ) is the task of assigning a semantic relationship between a pair of arguments---le and mikolov introduce paragraph vector to learn document representation from semantics of words | 0 |
sentiment analysis is the computational analysis of people ’ s feelings or beliefs expressed in texts such as emotions , opinions , attitudes , appraisals , etc . ( cite-p-11-3-3 )---sentiment analysis is the process of identifying and extracting subjective information using natural language processing ( nlp ) | 1 |
for feature building , we use word2vec pre-trained word embeddings---for creating the word embeddings , we used the tool word2vec 1 | 1 |
following the practice in learning to rank , we model document ranking in the pairwise style where the relevance information is in the form of preferences between pairs of documents with respect to individual queries---in this paper , we propose a novel method of reducing the size of translation model | 0 |
we use scikit-learn to implement the classifiers and accuracy scores to measure the predictability---we use several classifiers including logistic regression , random forest and adaboost implemented in scikit-learn | 1 |
we have presented a simple and effective method for learning the value of actions from reciprocal sentences---we present a simple and effective method for learning the value of actions from ranked pairs of textual action descriptions | 1 |
more precisely , we define several dependency trees exploitable by the partial tree kernel and compared them with stk over constituency trees---in particular , we define an efficient tree kernel derived from the partial tree kernel , suitable for encoding structural representation of comments into support vector machines | 1 |
these also include non-target entity specific features---due to the proposed non-target entity specific features | 1 |
lda is a probabilistic generative model that can be used to uncover the underlying semantic structure of a document collection---lda is a widely used topic model , which views the underlying document distribution as having a dirichlet prior | 1 |
semantic role labeling ( srl ) is the task of identifying semantic arguments of predicates in text---the semeval-2007 task 04 and semeval-2010 task 08 aimed at relations between nominals | 0 |
we train word embeddings using the continuous bag-of-words and skip-gram models described in mikolov et al as implemented in the open-source toolkit word2vec---in our experiments , we perform unsupervised learning of word-level embeddings using the word2vec tool 3 , which implements the continuous bag-of-words and skip-gram architectures for computing vector representations of words | 1 |
hiero is a hierarchical phrase-based statistical mt framework that generalizes phrase-based models by permitting phrases with gaps---the target-normalized hierarchical phrase-based model is based on a more general hierarchical phrase-based model | 1 |
word sense disambiguation ( wsd ) is the task to identify the intended sense of a word in a computational manner based on the context in which it appears ( cite-p-13-3-4 )---word sense disambiguation ( wsd ) is the task of determining the meaning of an ambiguous word in its context | 1 |
our usage of word embedding is in line with turian et al and yu et al , who study the effects of different clustering algorithms for pos tagging and named entity recognition---for example , turian et al have improved the performance of chunking and named entity recognition by using word embedding also as one of the features in their crf model | 1 |
the order in which phenomena are treated may therefore have a major impact on the resulting grammar---of the grammar , the resulting grammar is partially ( or even largely ) a product of the order in which phenomena are treated | 1 |
the model weights are automatically tuned using minimum error rate training---the model weights were trained using the minimum error rate training algorithm | 1 |
for english , we convert the ptb constituency trees to dependencies using the stanford dependency framework---we use the stanford dependency parser to parse the statement and identify the path connecting the content words in the parse tree | 1 |
we used the stanford parser to generate the grammatical structure of sentences---we pre-processed the data to add part-ofspeech tags and dependencies between words using the stanford parser | 1 |
for example , mihalcea and strapparava constructed the set of negative examples by using news title from reuters news , proverbs and british national corpus---we trained the l1-regularized logistic regression classifier implemented in liblinear | 0 |
shen et al proposed a string-to-dependency target language model to capture long distance word orders---typically , shen et al propose a string-todependency model , which integrates the targetside well-formed dependency structure into translation rules | 1 |
we train a 4-gram language model on the xinhua portion of the english gigaword corpus using the srilm toolkits with modified kneser-ney smoothing---we applied the noun clustering method of sun and korhonen to 2000 most frequent nouns in the bnc to obtain 200 common selectional preference classes | 0 |
work has also investigated whether scores on these dimensions correlate with language use---mikolov et al proposed a novel neural network model to train continuous vector representation for words | 0 |
blitzer et al investigate domain adaptation for sentiment classifiers , focusing on online reviews for different types of products---blitzer et al apply the structural correspondence learning algorithm to train a crossdomain sentiment classifier | 1 |
sun and xu enhanced the segmentation results by interpolating the statistics-based features derived from unlabeled data to a crfs model---sun and xu explored several statistical features derived from both unlabeled data to help improve character-based word segmentation | 1 |
the key idea is the following : we define a random walk on a graph over the items---the key idea is the tabulation of left-corner parsing , which captures the degree of center-embedding of a parse via its stack depth | 1 |
we then showed that by using these cross-lingual word clusters , we can significantly improve on direct transfer of discriminative models for both parsing and ner---and we show that features derived from these clusters significantly improve the accuracy of cross-lingual structure prediction | 1 |
one of the popular statistical machine translation paradigms is the phrase-based model---a popular statistical machine translation paradigms is the phrase-based model | 1 |
we used latent dirichlet allocation to perform the classification---we used latent dirichlet allocation to construct our topics | 1 |
rahman and ng used yago to inject knowledge attributes in mentions , but noticed that knowledge injection could be noisy---rahman and ng , 2011 ) used yago for similar purposes , but noticed that knowledge injection is often noisy | 1 |
in this work we propose the use of latent semantic analysis to induce a mdm from comparable corpora---in this work we exploit latent semantic analysis to automatically acquire a mdm from comparable corpora | 1 |
reinforcement learning is a machine learning technique that defines how an agent learns to take optimal actions in a dynamic environment so as to maximize a cumulative reward---reinforcement learning is a machine learning technique that defines how an agent learns to take optimal actions so as to maximise a cumulative reward | 1 |
while the majority of ccg parsers are chart-based , there has been some work on shift-reduce ccg parsing---while the majority of ccg parsers use chart-based approaches clark and curran , 2007 , there has been some work on developing shift-reduce parsers for ccg | 1 |
semantic role labeling ( srl ) is a major nlp task , providing a shallow sentence-level semantic analysis---model-refinement can dramatically decrease the bias introduced by ecoc , and the combined classifier is comparable to or even better than svm classifier in performance | 0 |
for language model , we use a trigram language model trained with the srilm toolkit on the english side of the training corpus---for the word-embedding based classifier , we use the glove pre-trained word embeddings | 0 |
crf is a well-known probabilistic framework for segmenting and labeling sequence data---it is a probabilistic framework proposed by for labeling and segmenting structured data , such as sequences , trees and lattices | 1 |
the neural embeddings were created using the word2vec software 3 accompanying---this baseline uses pre-trained word embeddings using word2vec cbow and fasttext | 1 |
in this paper , we used the decision list to solve the homophone problem---peng et al showed that better results can be achieved by global learning using a crf model | 0 |
we report decoding speed and bleu score , as measured by sacrebleu---we also measure overall performance with uncased bleu | 1 |
as monolingual baselines , we use the skip-gram and cbow methods of mikolov et al as implemented in the gensim package---we use the popular word2vec 1 tool proposed by mikolov et al to extract the vector representations of words | 1 |
relation extraction is the task of finding relationships between two entities from text---relation extraction is a crucial task in the field of natural language processing ( nlp ) | 1 |
pitler et al demonstrated that features developed to capture word polarity , verb classes and orientation , as well as some lexical features are strong indicator of the type of discourse relation---pitler et al show that pairs of words taken from sentences linked by discourse relations , as well as levin classes of verbs of the sentences and sentiment polarity information is useful for the prediction of implicit relations | 1 |
relation extraction is the task of predicting semantic relations over entities expressed in structured or semi-structured text---relation extraction ( re ) has been defined as the task of identifying a given set of semantic binary relations in text | 1 |
we therefore extend previous results to demonstrate the utility of this technique not only for a more semantically challenging task , but also a more complicated neural network architecture---with this transfer technique , demonstrating its effectiveness in a more complicated neural network architecture , and for a much more semantically challenging task | 1 |
it has already been proposed for phrase-based , hierarchical , and syntax-based systems---we evaluate our approaches on eight language pairs , with training data sizes ranging from 100k words to 8m words , and show improvements of up to + 4 . 3 bleu , surpassing phrase-based translation in nearly all settings | 0 |
we apply sri language modeling toolkit to train a 4-gram language model with kneser-ney smoothing---we used 5-gram models , estimated using the sri language modeling toolkit with modified kneser-ney smoothing | 1 |
we evaluated the system using bleu score on the test set---we report the mt performance using the original bleu metric | 1 |
our approach combines syntactic and word level similarity measures along with the unl based semantic similarity measures for finding similarity scores between sentences---for each question math-w-3-1-1-3 , let math-w-3-1-1-6 be the unstructured text and math-w-3-1-1-12 | 0 |
coreference resolution is the task of determining whether two or more noun phrases refer to the same entity in a text---coreference resolution is the problem of identifying which mentions ( i.e. , noun phrases ) refer to which real-world entities | 1 |
systems are tuned using pairwise ranking optimization on a different held-out opensubtitles set---feature weights are tuned using pairwise ranking optimization on the mt04 benchmark | 1 |
supervised machine learning methods including support vector machines are often used in sentiment analysis and shown to be very promising---supervised techniques have been proved promising and widely used in sentiment classification | 1 |
adding our copy action mechanism further increases this improvement ( +2.39 )---copy actions further improves this enhancement to reach + 2 . 39 | 1 |
bleu is a popular metric for evaluating statistical machine translation systems and fits our needs well---the bleu metric has deeply rooted in the machine translation community and is used in virtually every paper on machine translation methods | 1 |
rosa et al and mare膷ek et al applied a rule-based approach to ape of english-czech mt outputs on the morphological level---rosa et al and mare膷ek et al applied a rule-based approach to ape on the morphological level | 1 |
we integrated the transliteration extraction module into the giza++ word aligner and showed gains in alignment quality---we integrate a transliteration module into the giza + + word aligner and show that it improves word alignment quality | 1 |
wan et al , 2005 ) proposed a web person resolution system called webhawk , which extracted several attributes such as title , organization , email and phone number using patterns---al-sabbagh and girju described an approach of mining the web to build a da-to-msa lexicon | 0 |
auli et al propose a joint language and translation model , based on a recurrent neural network---the recurrent neural network lms of auli et al are primarily trained to predict target word sequences | 1 |
for this evaluation , we leverage rouge to address the relative quality of the generated summaries based on common ngram counts and longest common subsequence---similar to the evaluation for traditional summarization tasks , we use the rouge metrics to automatically evaluate the quality of produced summaries given the gold-standard reference news | 1 |
this paper proposes to directly optimize the search stage with a discriminative model based on latent structural svm---in this paper is a novel unified way to directly optimize the search phase of query | 1 |
relation extraction is a traditional information extraction task which aims at detecting and classifying semantic relations between entities in text ( cite-p-10-1-18 )---relation extraction is a fundamental task in information extraction | 1 |
semantic role labeling ( srl ) is a major nlp task , providing a shallow sentence-level semantic analysis---in this paper we explore the problem of making predictions for entities or entity-pairs unseen at training time ( and hence without a pre-learned row | 0 |
we are concerned with dependency-oriented morphosyntactic parsing of running text---we are concerned with grammar-based surface-syntactic analysis of running text | 1 |
feature weights were set with minimum error rate training on a development set using bleu as the objective function---for adjusting feature weights , the mert method was applied , optimizing the bleu-4 metric obtained on the development corpus | 1 |
learning algorithms used include maximum entropy , averaged perceptron , nave bayes , etc---models used include maximum entropy , averaged perceptron , naive bayes , etc | 1 |
prior work has reduced the size of smt phrase tables in order to improve efficiency without the loss of translation quality---johnson et al has shown that large portions of the phrase table can be removed without loss in translation quality | 1 |
we used a 5-gram language model with modified kneser-ney smoothing implemented using the srilm toolkit---since coreference resolution is a pervasive discourse phenomenon causing performance impediments in current ie systems , we considered a corpus of aligned english and romanian texts to identify coreferring expressions | 0 |
coreference resolution is a challenging task , that involves identification and clustering of noun phrases mentions that refer to the same real-world entity---coreference resolution is the task of determining which mentions in a text are used to refer to the same real-world entity | 1 |
we also extend cite-p-14-3-4 , which used a lexicon to learn bilingual word embeddings---as the pivot language , cite-p-15-4-4 learn multilingual word embeddings for many languages | 1 |
in this paper , we make a move to build a dialogue system for automatic diagnosis---in this paper , we propose a reinforcement learning based framework of dialogue system for automatic diagnosis | 1 |
the most well-known automatic evaluation metric in nlp is bleu for mt , based on n-gram matching precisions---we used the moses toolkit to build mt systems using various alignments | 0 |
in this work , we propose the dual tensor model , a neural architecture with which we explicitly model the asymmetry and capture the translation between unspecialized and specialized word embeddings via a pair of tensors---in this work , we propose the dual tensor model , a neural architecture that ( 1 ) models asymmetry more explicitly than existing models and ( 2 ) explicitly captures the translation of unspecialized distributional vectors into specialized embeddings | 1 |
dependency parsing is a basic technology for processing japanese and has been the subject of much research---dependency parsing is a very important nlp task and has wide usage in different tasks such as question answering , semantic parsing , information extraction and machine translation | 1 |
we initialize the embedding weights by the pre-trained word embeddings with 200 dimensional vectors---we use skip-gram with negative sampling for obtaining the word embeddings | 1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.