text stringlengths 82 736 | label int64 0 1 |
|---|---|
we also trained 5-gram language models using kenlm---gram language models were trained with lmplz | 1 |
experimental results on the nist mt-2003 chinese-english translation task show that our algorithm is at least 19 times faster in rule matching and is able to help to save 57 % of overall translation time over previous methods when using large fragment translation rules---chinese-english translation task shows that our algorithm is 19 times faster in rule matching and is able to save 57 % of overall translation time over previous methods when using large fragment translation rules | 1 |
for our second method , we developed a novel notion of feature coverage---for our second method , we develop the concept of feature coverage | 1 |
we propose using a principled way of incorporating both rater-comment and rater-author interactions simultaneously---bilingual lexicons are an important resource in multilingual natural language processing tasks such as statistical machine translation and cross-language information retrieval | 0 |
we evaluated the composite semi-supervised kpca model using data from the senseval-2 english lexical sample task---high quality word embeddings have been proven helpful in many nlp tasks | 0 |
to address this challenge , we propose a probabilistic approach for performing joint query annotation---to this end , we proposed a probabilistic approach for performing joint query annotation | 1 |
we built a 5-gram language model on the english side of europarl and used the kneser-ney smoothing method and srilm as the language model toolkit---the use of unsupervised word embeddings in various natural language processing tasks has received much attention | 0 |
a number of recent studies show that character sequence labeling is a simple but effective formulation of chinese word segmentation and name entity recognition for machine learning---recent studies show that character sequence labeling is an effective method of chinese word segmentation for machine learning | 1 |
sentiment classification is the task of detecting whether a textual item ( e.g. , a product review , a blog post , an editorial , etc . ) expresses a p ositive or a n egative opinion in general or about a given entity , e.g. , a product , a person , a political party , or a policy---system , the system participated in semeval 2016 community question ranking shared task for the arabic language | 0 |
we presented a maximum entropy model to extend the sentence compression methods described by knight and marcu---knight and marcu proposed a sentence compression method using a noisy-channel model | 1 |
we propose a grouping-based ordering framework that integrates local and global coherence concerns---we propose a grouping-based ordering framework that integrates both local and global coherence ; ( 3 ) | 1 |
sarcasm is a form of verbal irony that is intended to express contempt or ridicule---since sarcasm is a refined and indirect form of speech , its interpretation may be challenging for certain populations | 1 |
relation extraction is a core task in information extraction and natural language understanding---the x-lingual method uses unlabeled parallel sentences to induce cross-lingual word clusters as augmenting features for delexicalized dependency parser | 0 |
in this paper we present a new , publicly available corpus for context-dependent semantic parsing---in this paper we present a new corpus for context-dependent semantic parsing | 1 |
we also used word2vec to generate dense word vectors for all word types in our learning corpus---this goes beyond previous work on semantic parsing such as lu et al or zettlemoyer and collins which rely on unambiguous training data where every sentence is paired only with its meaning | 0 |
the weights of the different feature functions were optimised by means of minimum error rate training---the parameters for each phrase table were tuned separately using minimum error rate training | 1 |
the stanford parser can output typed semantic dependencies that conform to the stanford dependencies---the stanford dependency parser is used for extracting features from the dependency parse trees | 1 |
topic models , such as plsa and lda , have shown great success in discovering latent topics in text collections---generative models like lda and plsa have been proved to be very successful in modeling topics and other textual information in an unsupervised manner | 1 |
we introduce the novel task of automatically generating questions that are relevant to a text but do not appear in it---above , we propose the novel task of automatically suggesting comparative questions that are relevant to a given input | 1 |
event extraction and visualization are typically considered as two different tasks---modeling and visualization are considered as two disjoint tasks | 1 |
joty et al approach the document-level discourse parsing using a model trained by conditional random fields---joty et al approach the problem of textlevel discourse parsing using a model trained by conditional random fields | 1 |
we use the moses toolkit to train various statistical machine translation systems---as a baseline system for our experiments we use the syntax-based component of the moses toolkit | 1 |
kambhatla employs maximum entropy model to combine diverse lexical , syntactic and semantic features derived from the text in relation extraction---kambhatla employs maximum entropy models to combine diverse lexical , syntactic and semantic features derived from the text for relation extraction | 1 |
we performed significance testing using paired bootstrap resampling---we applied paired bootstrap resampling for a significance test | 1 |
light et al , medlock and briscoe , medlock , and szarvas ,---medlock and briscoe , vincze et al , and farkas et al , | 1 |
for representing words , we used 100 dimensional pre-trained glove embeddings---we used glove 10 to learn 300-dimensional word embeddings | 1 |
we adopt the iterative parameter mixing variation of the perceptron to scale to a large number of training examples---we use the idea of iterative parameter mixture to parallelize the training process | 1 |
we use the 300-dimensional pre-trained word2vec 3 word embeddings and compare the performance with that of glove 4 embeddings---for the neural models , we use 100-dimensional glove embeddings , pre-trained on wikipedia and gigaword | 1 |
ibm watson news explorer 3 gives a more analytical way to read news---ibm watson news explorer gives a more analytical way to read news through linked data | 1 |
snow et al demonstrated binary classification of hypernyms and non-hypernyms using wordnet as a source of training labels---snow et al utilize wordnet to learn dependency path patterns for extracting the hypernym relation from text | 1 |
we perform pre-training using the skip-gram nn architecture available in the word2vec 13 tool---to train the link embeddings , we use the speedy , skip-gram neural language model of mikolov et al via their toolkit word2vec | 1 |
interpretability and discriminative power are the two most basic requirements for an evaluation metric---interpretability and discriminative power are two basic requirements for a reasonable evaluation metric | 1 |
we used bootcat , a corpus building tool designed to collect data from the web , to collect our lm adaptation data---to build a corpus of mixed language documents , we used the bootcat tool seeded with words from a minority language | 1 |
we use a learned semantic lexicon to aid the construction of a smaller and more focused set of pcfg productions---therefore , our extension incorporates a learned lexicon to constrain the space of productions , thereby making the size of the pcfg tractable for complex | 1 |
we used minimum error rate training for tuning on the development set---the parameters of our mt system were tuned on a development corpus using minimum error rate training | 1 |
for english , we use the updated wsj with ontonotes-style annotations converted to stanford dependencies---as for ej translation , we use the stanford parser to obtain english abstraction trees | 1 |
in this study , our goal is to investigate how these two types of difficulty , namely “ answering questions ” and “ reading text , ” are correlated in rc---on the test corpus , we achieved a score of 0 . 465 with the simple english wikipedia system | 0 |
relation extraction is the task of finding relations between entities in text , which is useful for several tasks such as information extraction , summarization , and question answering ( cite-p-14-3-7 )---relation extraction ( re ) is a task of identifying typed relations between known entity mentions in a sentence | 1 |
collobert and weston , 2008 , proposed a multitask neural network trained jointly on the relevant tasks using weight-sharing---collobert and weston used convolutional neural networks in a multitask setting , where their model is trained jointly for multiple nlp tasks with shared weights | 1 |
we used the stanford parser to generate the grammatical structure of sentences---we use the stanford parser for obtaining all syntactic information | 1 |
ambiguity is the task of building up multiple alternative linguistic structures for a single input ( cite-p-13-1-8 )---ambiguity is the task of building up multiple alternative linguistic structures for a single input | 1 |
automated essay scoring utilizes the nlp techniques to automatically rate essays written for given prompts , namely , essay topics , in an educational setting---automated essay scoring utilizes natural language processing and machine learning techniques to automatically rate essays written for a target prompt | 1 |
ma and hovy presented a model that uses a convolutional network to compute representations for parts of words---ma and hovy proposed a lstm-cnns-crf model that utilizes convolutional neural networks to extract character-level features besides word-level features | 1 |
recently , dubey has proposed an approach that combines a probabilistic parser with a model of co-reference and discourse inference based on probabilistic logic---at the discourse level , dubey has proposed a model that combines an incremental parser with a probabilistic logic-based model of co-reference resolution | 1 |
we used a standard pbmt system built using moses toolkit---we used the moseschart decoder and the moses toolkit for tuning and decoding | 1 |
semantic role labeling ( srl ) is the task of automatically labeling predicates and arguments in a sentence with shallow semantic labels---semantic role labeling ( srl ) is the process of extracting simple event structures , i.e. , “ who ” did “ what ” to “ whom ” , “ when ” and “ where ” | 1 |
hoffmann et al present a multi-instance multi-label model for relation extraction through distant supervision---to alleviate the noise issue caused by distant supervision , riedel et al and hoffmann et al propose multi-instance learning mechanisms | 1 |
in this paper , we describe an rst-style text-level discourse parser based on a neural network model---in this paper , we propose a recursive model for discourse parsing that jointly models distributed representations | 1 |
compressing deep learning models is an active area of current research---compressing deep models into smaller networks has been an active area of research | 1 |
we use bleu 2 , ter 3 and meteor 4 , which are the most-widely used mt evaluation metrics---we use three common evaluation metrics including bleu , me-teor , and ter | 1 |
sentiwordnet is another popular lexical resource for opinion mining---sentiwordnet is a large lexicon for sentiment analysis and opinion mining applications | 1 |
the clear drawback of supervised methods is the need of training data : labeled data is expensive to obtain , and there is often a mismatch between the training data and the data the system will be applied to---our baseline russian-english system is a hierarchical phrase-based translation model as implemented in cdec | 0 |
wu et al compared machine learning methods for abbreviation detection---zeng et al exploit a convolutional neural network to extract lexical and sentence level features for relation classification | 0 |
we use a cws-oriented model modified from the skip-gram model to derive word embeddings---we first train a word2vec model on fr-wikipedia 11 to obtain non contextual word vectors | 1 |
for training and evaluating the itsg parser , we employ the penn wsj treebank---we have used penn tree bank parsing data with the standard split for training , development , and test | 1 |
the weights associated to feature functions are optimally combined using the minimum error rate training---the weights of these features are then learned using a discriminative training algorithm | 1 |
bilingual lexicons are an important resource in multilingual natural language processing tasks such as statistical machine translation and cross-language information retrieval---bilingual dictionaries of technical terms are important resources for many natural language processing tasks including statistical machine translation and cross-language information retrieval | 1 |
while each of these alternatives has some advantages over soundex , none is adaptable to alternative distance metrics---experimental results reflect that our method is effective | 0 |
furthermore , it is observed that opinion words and features themselves have relations in opinionated expressions too---it has been shown in that many feature and opinion word pairs have long range dependencies | 1 |
for interactive topic modeling , tandem anchors produce higher quality topics than single word anchors ( section 3 )---similarly , carbonell et al propose an mt method that needs no parallel text , but relies on a lightweight translation model utilising a fullform bilingual dictionary and a decoder for longrange context | 0 |
yamada and matsumoto proposed a deterministic classifierbased parser---yamada and matsumoto proposed a shift-reducelike odeterministic algorithm | 1 |
models were built and interpolated using srilm with modified kneser-ney smoothing and the default pruning settings---the language models were 5-gram models with kneser-ney smoothing built using kenlm | 1 |
we use a phrase-based translation system similar to moses---we develop translation models using the phrase-based moses smt system | 1 |
these models were first studied in the context of feed-forward networks , and later in the context of recurrent neural network models---the proposed neural models have a large number of variations , such as feed-forward networks , hierarchical models , recurrent neural networks , and recursive neural networks | 1 |
our word sense induction method is based on the effective procedure first presented by sch眉tze---for comparison with multi-prototype methods , we borrow the context-clustering idea from huang et al , which was first presented by sch眉tze | 1 |
approaches like and use a dictionary to check if a decipherment is useful---approaches like hart and olson use a dictionary to check if a decipherment is useful | 1 |
in addition , pitler and nenkova presented a comparison of texts in terms of difficulty by using an svm---pitler and nenkova used the same features to evaluate how well a text is written | 1 |
relation extraction is a fundamental task in information extraction---relation extraction ( re ) is the task of recognizing the assertion of a particular relationship between two or more entities in text | 1 |
an english 5-gram language model is trained using kenlm on the gigaword corpus---in this task , we used conditional random fields | 0 |
next , we use wordnet to identify synonyms of the content words---next we group these words using wordnet to obtain more general concepts | 1 |
semantic role labeling ( srl ) is the task of identifying semantic arguments of predicates in text---nakagawa et al , 2010 ) introduced an approach based on crfs with hidden variables with very good performance | 0 |
the 5-gram kneser-ney smoothed language models were trained by srilm , with kenlm used at runtime---the process of identifying the correct meaning , or sense of a word in context , is known as word sense disambiguation ( wsd ) | 0 |
luong et al utilized the morpheme segments produced by morfessor and constructed morpheme trees for words to learn morphologically-aware word embeddings by the recursive neural network---luong et al segment words using morfessor , and use recursive neural networks to build word embeddings from morph embeddings | 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 clustering a sequence of textual entity mentions into a set of maximal non-overlapping clusters , such that mentions in a cluster refer to the same discourse entity | 1 |
we use the glove word vector representations of dimension 300---we attempt to represent the cross-linguistic similarities that exist in the consonant inventories of the world ¡¯ s languages through a bipartite network | 0 |
we use the latest version of meteor that finds alignments between sentences based on exact , stem , synonym and paraphrase matches between words and phrases---we use the latest version of meteor that find alignments between sentences based on exact , stem , synonym and paraphrase matches between words and phrases | 1 |
twitter consists of a massive number of posts on a wide range of subjects , making it very interesting to extract information and sentiments from them---twitter is a popular microblogging service which provides real-time information on events happening across the world | 1 |
in recent years , many researchers build alignment links with bilingual corpora---in this phase were constrained based on the messages from the first phase | 0 |
in this paper , we present a novel precedence reordering approach based on a dependency parser---we introduce a novel precedence reordering approach based on a dependency parser | 1 |
we tune model weights using minimum error rate training on the wmt 2008 test data---we perform the mert training to tune the optimal feature weights on the development set | 1 |
the key to this success is the combination of two different views as in co-training ( cite-p-11-1-0 ) : an information extraction technique with fine features for high precision and an information retrieval technique with coarse features for high recall---the key for success is the use of unlabeled data with svd , a combination of kernels and svm | 1 |
language models were built with srilm , modified kneser-ney smoothing , default pruning , and order 5---the srilm toolkit was used for training the language models using kneser-ney smoothing | 1 |
these experiments are based on comparisons of performance using propbanked wsj data and propbanked brown corpus data---experiments are based on comparisons of performance using propbanked wsj data and propbanked brown corpus data | 1 |
it consists of more than one composition functions , and we model the adaptive sentiment propagations as distributions over these composition functions---and we model the adaptive sentiment propagations as learning distributions over these composition functions | 1 |
we follow the standard machine translation procedure of evaluation , measuring bleu for every system---we use mira to tune the parameters of the system to maximize bleu | 1 |
we first train a word2vec model on fr-wikipedia 11 to obtain non contextual word vectors---word sense disambiguation has been an open problem in computational linguistics | 0 |
for two question sets , a context for the target word is provided , and we examine a number of word similarity measures that exploit this context---for word similarity measures , we compare the results of several different measures and frequency estimates | 1 |
relation extraction ( re ) is the task of extracting semantic relationships between entities in text---for our purpose we use word2vec embeddings trained on a google news dataset and find the pairwise cosine distances for all words | 0 |
in this paper we present a sample ensemble parse assessment ( sepa ) algorithm for detecting parse quality---in this paper we introduced sepa , a novel algorithm for assessing parse quality | 1 |
the task of inducing hierarchical syntactic structure from observed yields alone has received a great deal of attention---the task of statistically inducing hierarchical syntactic structure over unannotated sentences of natural language has received a great deal of attention | 1 |
they have been successful in explaining a wide range of behavioral data -examples include lexical priming , deep dyslexia , text comprehension , synonym selection , and human similarity judgments---these models have proven successful in tasks relying on meaning relatedness , such as synonymy detection , word sense discrimination , or even measuring phrase plausibility | 1 |
the hierarchical model is built on a weighted synchronous contextfree grammar---the model can be formalized as a synchronous context-free grammar | 1 |
turian et al applied this method to both named entity recognition and text chunking---turian et al applied word embeddings to chunking and named entity recognition | 1 |
we build a baseline error correction system , using the moses smt system---typical language features are label en-coders and word2vec vectors | 0 |
the language is a form of modal propositional logic---language is a dynamic system , constantly evolving and adapting to the needs of its users and their environment ( cite-p-15-1-0 ) | 1 |
a major motivation for unsupervised morphological analysis is to reduce the sparse data problem in under-resourced languages---one of the motivations for unsupervised morphological analysis is to reduce data sparsity in downstream applications | 1 |
a knowledge base is a large repository of facts that are mainly represented as rdf triples , each of which consists of a subject , a predicate ( relationship ) , and an object---a knowledge base consists of a set of entities , and each entity can have a variation list 2 | 1 |
as the encoder for text we consider convolutional neural networks , gated recurrent units , and long short-term memory networks---for sequence modeling in all three components , we use the long short-term memory recurrent neural network | 1 |
choudhury et al describe a supervised noisy channel model using hmms for sms normalization---choudhury et al developed a hidden markov model using hand annotated training data | 1 |
we measured the overall translation quality with 4-gram bleu , which was computed on tokenized and lowercased data for all systems---we measure the overall translation quality using 4-gram bleu , which is computed on tokenized and lowercased data for all systems | 1 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.