text stringlengths 82 736 | label int64 0 1 |
|---|---|
zeng et al , 2014 , exploited a convolutional deep neural network to extract lexical and sentence level features---zeng et al proposed a deep convolutional neural network with softmax classification , extracting lexical and sentence level features | 1 |
we incorporate two domain-specific resources , i.e. , umls and a list of tumor names---and utilizes two domain-specific resources ( umls and a tumor name list ) | 1 |
we used the phrase-based smt in moses 5 for the translation experiments---for phrase-based smt translation , we used the moses decoder and its support training scripts | 1 |
sentiment analysis is a research area where does a computational analysis of people ’ s feelings or beliefs expressed in texts such as emotions , opinions , attitudes , appraisals , etc . ( cite-p-12-1-3 )---in order to establish the long dependencies easily and overcome the disadvantage of the approximate inference , krishnan and manning propose a two-stage approach using conditional random fields with extract inference | 0 |
for english , we use the pre-trained glove vectors---we use the pre-trained glove vectors to initialize word embeddings | 1 |
models that learn semantic representations from both linguistic and perceptual input outperform text-only models in many contexts and better reflect human concept acquisition---models that learn semantic concept representations from both linguistic and perceptual input were originally motivated by parallels with human concept acquisition , and evidence that many concepts are grounded in the perceptual system | 1 |
the word embeddings can provide word vector representation that captures semantic and syntactic information of words---word embedding thus have powerful capability to capture both semantic and syntactic variations of words | 1 |
four training and testing corpora were used in the first bakeoff , including the academia sinica corpus , the penn chinese treebank corpus , the hong kong city university corpus , and the peking university corpus---four training and testing corpora were used in the first bakeoff , including the academia sinica corpus , the penn chinese treebank corpus , the hong kong city university corpus and the peking university corpus | 1 |
we use the skip-gram model , trained to predict context tags for each word---we use the word2vec tool to pre-train the word embeddings | 1 |
we further used a 5-gram language model trained using the srilm toolkit with modified kneser-ney smoothing---we used the srilm toolkit to build unpruned 5-gram models using interpolated modified kneser-ney smoothing | 1 |
phrasebased smt models are tuned using minimum error rate training---the log-linear feature weights are tuned with minimum error rate training on bleu | 1 |
for evaluation , we use the dataset from the semeval-2007 lexical substitution task---in practical terms , we will use a paraphrase ranking task derived from the semeval 2007 lexical substitution task | 1 |
we used glove 10 to learn 300-dimensional word embeddings---semantic characterization of the inferred classes ¨c a prerequisite for using them in nlp tasks in an informed way | 0 |
this architecture is based on a two-layer convolutional neural network ensembled with a final long short-term memory neural network as in---the underlying model used is a long shortterm memory recurrent neural network in a bidirectional configuration | 1 |
semantic parsing is the task of translating natural language utterances into a machine-interpretable meaning representation---semantic parsing is the task of mapping a natural language ( nl ) sentence into a complete , formal meaning representation ( mr ) which a computer program can execute to perform some task , like answering database queries or controlling a robot | 1 |
the model parameters of word embedding are initialized using word2vec---typical language features are label en-coders and word2vec vectors | 1 |
in this paper we presented a word sense disambiguation based system for multilingual lexical substitution---in this paper , we present our system that was applied to the cross lingual substitution | 1 |
dependency parsing is a core task in nlp , and it is widely used by many applications such as information extraction , question answering , and machine translation---therefore , dependency parsing is a potential “ sweet spot ” that deserves investigation | 1 |
to obtain their corresponding weights , we adapted the minimum-error-rate training algorithm to train the outside-layer model---we adapt the minimum error rate training algorithm to estimate parameters for each member model in co-decoding | 1 |
choi et al , showed how to enhance chinese-english verb alignments by exploring predicate-argument structure alignment using parallel propbanks---word sense disambiguation ( wsd ) is a task to identify the intended sense of a word based on its context | 0 |
the language models are 4-grams with modified kneser-ney smoothing which have been trained with the srilm toolkit---n-gram features were based on language models of order 5 , built with the srilm toolkit on monolingual training material from the europarl and the news corpora | 1 |
in particular , we use measures such as translation model entropy , inspired by koehn et al---we adopt the idea of translation model entropy from koehn et al | 1 |
plagiarism is a major issue in science and education---plagiarism is a problem of primary concern among publishers , scientists , teachers ( cite-p-21-1-7 ) | 1 |
the output of our experiments was evaluated using two metrics , bleu , and lexical accuracy---as indicated in mikolov et al , word vectors obtained from deep learning neural net models exhibit linguistic regularities , such as additive compositionality | 0 |
fern谩ndez et al found that participants with ad had an increased number of total fixations , first-pass fixations , and second-pass fixations---fern谩ndez et al found that participants with ad had an increased number of fixations and regressions , and also skipped more words than healthy controls | 1 |
finkel and manning demonstrate the hierarchical bayesian extension of this where domain-specific models draw from a general base distribution---we use the ontonotes datasets from the conll 2011 shared task 6 , only for training the out-of-the-box system | 0 |
we make use of the english dependency treebank , developed on the computational paninian grammar model , for this work---we have modeled the simple parser on the paninian grammatical model which provides a dependency grammar framework | 1 |
we show that , surprisingly , dynamic programming is in fact possible for many shift-reduce parsers , by merging “ equivalent ” stacks based on feature values---for a large class of modern shift-reduce parsers , dynamic programming is in fact possible and runs in polynomial time | 1 |
conversely , ganchev et al developed a technique to optimize the more desirable reverse property of the word types having a sparse posterior distribution over tags---ganchev et al propose postcat which uses posterior regularization to enforce posterior agreement between the two models | 1 |
currently , studies are mainly concerned with the binary evaluation of humor , whether it is funny or not---however , most recent studies are concerned with a binary perspective over humor | 1 |
when parsers are trained on ptb , we use the stanford pos tagger---we used the srilm toolkit and kneser-ney discounting for estimating 5-grams lms | 0 |
we evaluate text generated from gold mr graphs using the well-known bleu measure---we evaluate our models using the standard bleu metric 2 on the detokenized translations of the test set | 1 |
the wordnet domains resource assigns domain labels to synsets in wordnet---wordnet domains was created by extending the princeton wordnet with domains labels | 1 |
we discuss how the two approaches perform relative to each other , and how characteristics of the corpus affect the suitability of different approaches and their outcomes---and we discuss how they perform relative to each other , and how characteristics of the corpus affect outcomes and the suitability of the two approaches | 1 |
word sense distributions are usually skewed---since sense distributions are usually skewed | 1 |
the language model was constructed using the srilm toolkit with interpolated kneser-ney discounting---we use the mallet implementation of conditional random fields | 0 |
a standard sri 5-gram language model is estimated from monolingual data---we used the srilm toolkit to create 5-gram language models with interpolated modified kneser-ney discounting | 0 |
however , the datasets used for evaluation have limitations in both scale and diversity---in recent work , the datasets employed in their evaluation are small and lack diversity | 1 |
we also used pre-trained word embeddings , including glove and 300d fasttext vectors---we used glove vectors trained on common crawl 840b 4 with 300 dimensions as fixed word embeddings | 1 |
coreference resolution is a field in which major progress has been made in the last decade---coreference resolution is the next step on the way towards discourse understanding | 1 |
in this paper , we describe a system that is able to learn context-sensitive features within the sentences---in this paper , we describe our participation in the multilingual sts task | 1 |
we introduce tiered clustering , a mixture model capable of accounting for varying degrees of shared ( context-independent ) feature structure , and demonstrate its applicability to inferring distributed representations of word meaning---erkan and radev proposed lexpagerank to compute the sentence saliency based on the concept of eigenvector centrality | 0 |
ng et al proposed that rather than focusing on just adjective-noun relationships , the subject-verb and verb-object relationships should also be considered for polarity classification---we used srilm -sri language modeling toolkit to train several character models | 0 |
the accuracy was measured using the bleu score and the string edit distance by comparing the generated sentences with the original sentences---the system output is evaluated using the meteor and bleu scores computed against a single reference sentence | 1 |
optimisation of the parameters is done using the sgd-based adam method and we perform gradient clipping to prevent exploding gradients---we estimated 5-gram language models using the sri toolkit with modified kneser-ney smoothing | 0 |
pronouns are resolved using a rule-based reimplementation of the cogniac algorithm and sentences are lemmatized and chunked using the cass chunker---the words are lemmatized using an augmented version of the scol toolset and sentences are chunked using the cass chunker | 1 |
morphological disambiguation is a well studied problem in the literature , but lstm-based contributions are still relatively scarce---morphological disambiguation is the process of assigning one set of morphological features to each individual word in a text | 1 |
in all cases , we used the implementations from the scikitlearn machine learning library---we used the first-stage pcfg parser of charniak and johnson for english and bitpar for german | 0 |
additionally , coreference resolution is a pervasive problem in nlp and many nlp applications could benefit from an effective coreference resolver that can be easily configured and customized---coreference resolution is a challenging task , that involves identification and clustering of noun phrases mentions that refer to the same real-world entity | 1 |
also , a number of semi-supervised word aligners have been proposed---a number of semi-supervised word aligners are proposed | 1 |
sarcasm is often used by individuals to express opinions on complex matters and regarding specific targets ( cite-p-15-1-6 )---the spelling normalisation component is a character-based statistical machine translation system implemented with the moses toolkit | 0 |
llu铆s et al use a joint arcfactored model that predicts full syntactic paths along with predicate-argument structures via dual decomposition---llu铆s et al introduce a joint arc-factored model for parsing syntactic and semantic dependencies , using dualdecomposition to maximize agreement between the models | 1 |
as embedding vectors , we used the publicly available representations obtained from the word2vec cbow model---in the remainder of this paper , sec . 2 illustrates the related work , sec . 3 introduces the complexity of learning entailments from examples , sec . 4 describes our models , sec . 6 shows the experimental results | 0 |
sentiment analysis is a ‘ suitcase ’ research problem that requires tackling many nlp subtasks , e.g. , aspect extraction ( cite-p-26-3-15 ) , named entity recognition ( cite-p-26-3-6 ) , concept extraction ( cite-p-26-3-20 ) , sarcasm detection ( cite-p-26-3-16 ) , personality recognition ( cite-p-26-3-7 ) , and more---we propose and explore the use of attention in a deep learning architecture to simulate the semantic priming mechanism | 0 |
the tuning step used minimum error rate training---the minimum error rate training was used to tune the feature weights | 1 |
this dataset has been used for evaluations in various semantic parsing works---it was used in many previous research efforts on semantic parsing | 1 |
we employ a neural method , specifically the continuous bag-of-words model to learn high-quality vector representations for words---we use a count-based distributional semantics model and the continuous bag-of-words model to learn word vectors | 1 |
collobert et al proposed cnn architecture that can be applied to various nlp tasks , such as pos tagging , chunking , named entity recognition and semantic role labeling---collobert et al , 2011 ) used word embeddings for pos tagging , named entity recognition and semantic role labeling | 1 |
our trigram word language model was trained on the target side of the training corpus using the srilm toolkit with modified kneser-ney smoothing---the cleansed part of our training dataset , which consists of more than 55,000 words , is fed into srilm to compile a bidirectional trigram lm by employing the kneser-ney smoothing algorithm | 1 |
empirically , s-lstm can give effective sentence encoding after 3 – 6 recurrent steps---however , s-lstm models hierarchical encoding of sentence structure as a recurrent state | 1 |
ideally , it can be estimated by using the forward-backward algorithm recursively for the first-order or second-order hmms---ideally the third item can be estimated by the forward-backward algorithm recursively for the firstorder or second-order hmms | 1 |
besides phrase-based machine translation systems , syntax-based systems have become widely used because of their ability to handle non-local reordering---phrase-based models have been widely used in practical machine translation systems due to their effectiveness , simplicity , and applicability | 1 |
the latent dirichlet allocation is a topic model that is assumed to provide useful information for particular subtasks---latent dirichlet allocation is a topic modeling framework that is often used for text classification | 1 |
since the training data is imbalanced , we specifically designed a two-step classifier to address subtask a---to learn the topics we use latent dirichlet allocation | 0 |
sch眉tze , 1998 ) utilized second order context vectors that represent the context of a target word to be discriminated by taking the average of the first order vectors associated with the unigrams that occur in that context---however , reference resolution remains a challenging problem , partly due to limited speech and language processing capabilities | 0 |
blitzer et al employ the structural correspondence learning algorithm for sentiment domain adaptation---blitzer et al used the structural correspondence learning algorithm with mutual information | 1 |
we used the svm implementation provided within scikit-learn---for training our system classifier , we have used scikit-learn | 1 |
multiword expressions are defined as idiosyncratic interpretations that cross word boundaries or spaces---mutiword terms defined as idiosyncratic interpretations cross word boundaries | 1 |
relation extraction ( re ) is the process of generating structured relation knowledge from unstructured natural language texts---we also use a 4-gram language model trained using srilm with kneser-ney smoothing | 0 |
mcgough et al proposed an approach to build a web-based testing system with the facility of dynamic question generation---mcgough et al proposed an approach to build a web-based testing system with the facility of dynamic qg | 1 |
we also present a fast approximate method for performing gcca and approximately recover the objective of ( cite-p-13-3-27 ) while accounting for missing values---we present an algorithm for fast approximate computation of gcca , which when coupled with methods for handling missing values | 1 |
to solve this problem , we employed supervised machine learning techniques exploiting a rich feature set---we proposed a supervised machine learning technique that employs a rich feature set | 1 |
as a countbased baseline , we use modified kneser-ney as implemented in kenlm---for all the systems we train , we build n-gram language model with modified kneserney smoothing using kenlm | 1 |
then the best linearizations compatible with the relative order are selected by log-linear models---then the best linearization for each subtree is selected by the log-linear model | 1 |
we extract dependency structures from the penn treebank using the head rules of yamada and matsumoto---we use the english penn treebank to evaluate our model implementations and yamada and matsumoto head rules are used to extract dependency trees | 1 |
the lms are build using the srilm language modelling toolkit with modified kneserney discounting and interpolation---in this work , we propose a general graph representation for automatically extracting structured features from tokens and prior annotations | 0 |
very recently , by representing gr analysis using general directed dependency graphs , sun et al and zhang et al showed that considerably good gr structures can be directly obtained using data-driven , transition-based parsing techniques---by encoding grs as directed graphs over words , sun et al and zhang et al showed that the data-driven , transition-based approach can be applied to build chinese gr structures with very promising results | 1 |
semantic parsing is the task of mapping natural language to machine interpretable meaning representations---semantic parsing is the task of mapping natural language utterances to machine interpretable meaning representations | 1 |
sentiment classification has seen a great deal of attention---sentiment detection and classification has received considerable attention | 1 |
in their system , the division is managed with parameters that control how many categories the parser ’ s chart is seeded with---in this scenario , we could simply lift the sets of relevant sentences from each document | 0 |
lin and demner-fushman clustered medline citations based on the occurrence of specific mentions of interventions in the document abstracts---lin and demner-fushman grouped medline citations into clusters based on interventions extracted from the document abstracts | 1 |
in order to reduce the vocabulary size , we apply byte pair encoding---in addition to the attention model , we use byte pair encoding in the preprocessing step | 1 |
question answering ( qa ) is a well-studied problem in nlp which focuses on answering questions using some structured or unstructured sources of knowledge---question answering ( qa ) is a challenging task that draws upon many aspects of nlp | 1 |
the algorithm we have proposed is language independent : it exploits a maximum entropy letters model trained over the known words observed in the corpus and the distribution of the unknown words in known tag contexts , through iterative approximation---independent : it exploits a maximum entropy letters model trained over the known words observed in the corpus and the distribution of the unknown words in known tag contexts , through iterative approximation | 1 |
anchor verbs can be selected by focusing on their arguments---verbs which take topical nouns can be candidates for anchor verbs | 1 |
we train skip-gram word embeddings with the word2vec toolkit 1 on a large amount of twitter text data---turney and littman determined the semantic orientation of a target word t by comparing its association with two seed sets of manually crafted target words | 0 |
we pre-train the word embedding via word2vec on the whole dataset---our dataset of informal medical narratives consists of verbal autopsy reports from the million death study , a program that collects vas in india that cover adult , child , and neonatal deaths | 0 |
hsueh and moore then trained a maximum entropy classifier to recognize this single da class , using a variety of lexical , prosodic , dialogue act and conversational topic features---hsueh and moore , 2007 ) then trained a maximum entropy classifier to recognize this single da class , using a variety of lexical , prosodic , da and conversational topic features | 1 |
images from google have been shown to yield representations that are competitive in quality compared to alternative resources---we train a 4-gram language model on the xinhua portion of the gigaword corpus using the sri language toolkit with modified kneser-ney smoothing | 0 |
feature weights were set with minimum error rate training on a tuning set using bleu as the objective function---system tuning was carried out using both k-best mira and minimum error rate training on the held-out development set | 1 |
attitude predictions are used to construct a signed network representation of the discussion thread---discussants is identified , this information is then used to construct a signed network representation of the discussion thread | 1 |
we further used a 5-gram language model trained using the srilm toolkit with modified kneser-ney smoothing---a 5-gram language model was created with the sri language modeling toolkit and trained using the gigaword corpus and english sentences from the parallel data | 1 |
we present our machine learning system which utilizes lexical , syntactical and semantic based feature sets---we present a machine learning based system for extraction of drug-drug interactions , using lexical , syntactic and semantic | 1 |
we obtain these dependency constructions by implementing a distantly supervised pattern extraction approach---for scalability , our systems uses distantly supervised data to train a relation extraction model | 1 |
active learning is a framework that makes it possible to efficiently train statistical models by selecting informative examples from a pool of unlabeled data---li et al , li and zhou , hatori et al , and ma et al present systems that jointly model chinese pos tagging and dependency parsing | 0 |
the grammar consists of head-dependent relations between words and can be learned automatically from a raw corpus using the reestimation algorithm which is also introduced in this paper---each grammar consists of a set of rules evaluated in a leftto-right fashion over the input annotations , with multiple grammars cascaded together and evaluated bottom-up | 1 |
in this research , we have build a wikification corpus for advancing japanese entity linking---in this research , we build a japanese wikification corpus in which mentions in japanese | 1 |
we use conditional random fields , a popular approach to solve sequence labeling problems---we use conditional random field sequence labeling as described in | 1 |
the obtained triple translation model is also used for collocation translation extraction---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 |
ner is a task to identify names in texts and to assign names with particular types ( cite-p-12-3-17 , cite-p-12-3-19 , cite-p-12-3-18 , cite-p-12-3-2 )---this problem can be cast as an instance of synchronous itg parsing | 0 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.