Filters








467 Hits in 4.6 sec

A neural-network architecture for syntax analysis

Chun-Hsien Chen, V. Honavar
1999 IEEE Transactions on Neural Networks  
Index Terms-Lexical analysis, modular neural networks, neural associative processing, neural associative processor, neural parser, neural symbolic processing, parsing, syntax analysis.  ...  Each of these modules takes advantage of parallel content-based pattern matching using a neural associative memory.  ...  His research interests include artificial intelligence and neural networks. His thesis research emphasizes neural architectures for knowledge representation and inference.  ... 
doi:10.1109/72.737497 pmid:18252507 fatcat:76puh6lgtbhbfcoab5pb33qsfe

Efficient Structured Inference for Transition-Based Parsing with Neural Networks and Error States

Ashish Vaswani, Kenji Sagae
2016 Transactions of the Association for Computational Linguistics  
Using neural networks for our local classifiers, our approach achieves 93.61% accuracy for transition-based dependency parsing in English.  ...  In this paper, we propose a new approach for approximate structured inference for transition-based parsing that produces scores suitable for global scoring using local models.  ...  Trained without external resources or pre-trained embeddings, our neural network (NN) dependency parser outperforms the NN transition-based dependency parser from Chen and Manning (2014) , which uses  ... 
doi:10.1162/tacl_a_00092 fatcat:j6v6gxsyx5eyxmvpcb4xqcweha

Assessing the Use of Prosody in Constituency Parsing of Imperfect Transcripts [article]

Trang Tran, Mari Ostendorf
2021 arXiv   pre-print
The neural parser is based on a sentence encoder that leverages word vectors contextualized with prosodic features, jointly learning prosodic feature extraction with parsing.  ...  We assess the utility of the prosody in parsing on imperfect transcripts, i.e. transcripts with automatic speech recognition (ASR) errors, by applying the parser in an N-best reranking framework.  ...  The acoustic-prosodic features si are learned via a convolutional neural network (CNN) from energy (E) and pitch (f0) contours as described in [10] .  ... 
arXiv:2106.07794v1 fatcat:bhgiarjcefe63fgym42yesi4rm

Improving Classifiers for Semantic Annotation of Software Requirements with Elaborate Syntactic Structure

Yeongsu Kim, Seungwoo Lee, Markus Dollmann, Michaela Geierhos
2018 International Journal of Advanced Science and Technology  
To improve our classifier, we carefully designed syntactic features extracted by constituency and dependency parsers.  ...  Even with a small dataset and a large number of classes, our proposed classifier records an accuracy of 0.75, which outperforms the previous model, REaCT.  ...  Acknowledgments This paper is a revised and expanded version of a paper entitled "Semantic Annotation of Software Requirements with Language Frame" presented at the 14 th 2017 International Interdisciplinary  ... 
doi:10.14257/ijast.2018.112.12 fatcat:murl3jl2kba6njjl445m2ulliq

CCG Supertagging with a Recurrent Neural Network

Wenduan Xu, Michael Auli, Stephen Clark
2015 Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)  
Recent work on supertagging using a feedforward neural network achieved significant improvements for CCG supertagging and parsing (Lewis and Steedman, 2014).  ...  In this paper, we show how directly capturing sequence information using a recurrent neural network leads to further accuracy improvements for both supertagging (up to 1.9%) and parsing (up to 1% F1),  ...  We introduce a recurrent neural network-based (RNN) supertagging model to tackle all the above problems, with an emphasis on the third one.  ... 
doi:10.3115/v1/p15-2041 dblp:conf/acl/XuAC15 fatcat:knfkb53zpfebxcuzbuhcwx27vq

Two Local Models for Neural Constituent Parsing [article]

Zhiyang Teng, Yue Zhang
2018 arXiv   pre-print
Such features have been a key to the success of state-of-the-art statistical parsers.  ...  Non-local features have been exploited by syntactic parsers for capturing dependencies between sub output structures.  ...  Span Model Given an unlabeled binarized tree T ub for the sentence S, S = w 0 , w 1 . . . w n−1 , the span model trains a neural network model P (Y [i,j] |S, Θ) to distinguish constituent spans from  ... 
arXiv:1808.04850v2 fatcat:j4de5mfyivb6zhmodbtnc4xuru

Expected F-Measure Training for Shift-Reduce Parsing with Recurrent Neural Networks

Wenduan Xu, Michael Auli, Stephen Clark
2016 Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies  
We present expected F-measure training for shift-reduce parsing with RNNs, which enables the learning of a global parsing model optimized for sentence-level F1.  ...  We apply the model to CCG parsing, where it improves over a strong greedy RNN baseline, by 1.47% F1, yielding state-of-the-art results for shiftreduce CCG parsing.  ...  In this paper, we present a global neural network parsing model, optimized for a task-specific loss based on expected F-measure.  ... 
doi:10.18653/v1/n16-1025 dblp:conf/naacl/XuAC16 fatcat:i5uqhjbihnhnzh5z37riix7va4

Wide coverage natural language processing using kernel methods and neural networks for structured data

Sauro Menchetti, Fabrizio Costa, Paolo Frasconi, Massimiliano Pontil
2005 Pattern Recognition Letters  
Convolution kernels and recursive neural networks are both suitable approaches for supervised learning when the input is a discrete structure like a labeled tree or graph.  ...  In both problems, the learning task consists in choosing the best alternative tree in a set of candidates.  ...  Recursive neural networks preference model We implement U(x) by a neural network having a single linear output.  ... 
doi:10.1016/j.patrec.2005.03.011 fatcat:e4k3ynjycvc7je3xxqarle5nm4

Grammar inference algorithms and applications in software engineering

Marjan Mernik, Dejan Hrncic, Barrett R. Bryant, Alan P. Sprague, Jeff Gray, Qichao Liu, Faizan Javed
2009 2009 XXII International Symposium on Information, Communication and Automation Technologies  
In the case when there is a lack of domain experts, grammar inference can be applied.  ...  ACKNOWLEDGMENT This material is based upon work partially supported by the National Science Foundation under Grant No. 0811630.  ...  ., neural networks, structured data and patterns).  ... 
doi:10.1109/icat.2009.5348441 dblp:conf/icatech/MernikHBSGLJ09 fatcat:zjfjaxqwn5h67mvsmi7bqr22fe

PELESent: Cross-domain polarity classification using distant supervision [article]

Edilson A. Corrêa Jr, Vanessa Q. Marinho, Leandro B. dos Santos, Thales F. C. Bertaglia, Marcos V. Treviso, Henrico B. Brum
2017 arXiv   pre-print
Our methods obtained very competitive results in five annotated corpora from mixed domains (Twitter and product reviews), which proves the domain-independent property of such approach.  ...  The main goal of this task is to classify the polarity of a message.  ...  The corpora were manually annotated in positive and negative, and used to evaluate stream based sentiment analysis systems. b) Buscapé [26] : This dataset is formed by product reviews extracted from Buscapé  ... 
arXiv:1707.02657v1 fatcat:xkh4ykfmuvbhnn2lstfkmartrm

A Fast Unified Model for Parsing and Sentence Understanding

Samuel R. Bowman, Jon Gauthier, Abhinav Rastogi, Raghav Gupta, Christopher D. Manning, Christopher Potts
2016 Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)  
We address these issues by introducing the Stack-augmented Parser-Interpreter Neural Network (SPINN), which combines parsing and interpretation within a single tree-sequence hybrid model by integrating  ...  Tree-structured neural networks exploit valuable syntactic parse information as they interpret the meanings of sentences.  ...  Acknowledgments We acknowledge financial support from a Google Faculty Research Award, the Stanford Data Science Initiative, and the National Science Foundation under grant nos.  ... 
doi:10.18653/v1/p16-1139 dblp:conf/acl/BowmanGRGMP16 fatcat:43j7ghruuzcmta3gvr664fzlh4

NLP-Based Subject with Emotions Joint Analytics for Epidemic Articles

Woo Hyun Park, Isma Farah Siddiqui, Dong Ryeol Shin, Nawab Muhammad Faseeh Qureshi
2022 Computers Materials & Continua  
The first is a subject and non-linear emotional module, which extracts topics from the data and supplements them with emotional figures.  ...  The accuracy and other model measurements, such as the F1, area under the curve, and recall are evaluated based on an article on Middle East respiratory syndrome.  ...  Mean square error (MSE) was used based on the product of the emotion and subject vectors.  ... 
doi:10.32604/cmc.2022.028241 fatcat:mgbxipxo7zfwzaawexc5ulcb7a

A Fast Unified Model for Parsing and Sentence Understanding [article]

Samuel R. Bowman, Jon Gauthier, Abhinav Rastogi, Raghav Gupta, Christopher D. Manning, Christopher Potts
2016 arXiv   pre-print
We address these issues by introducing the Stack-augmented Parser-Interpreter Neural Network (SPINN), which combines parsing and interpretation within a single tree-sequence hybrid model by integrating  ...  Tree-structured neural networks exploit valuable syntactic parse information as they interpret the meanings of sentences.  ...  Acknowledgments We acknowledge financial support from a Google Faculty Research Award, the Stanford Data Science Initiative, and the National Science Foundation under grant nos.  ... 
arXiv:1603.06021v3 fatcat:oejlz7jolrhc5gg2yeljl23rcm

Does constituency analysis enhance domain-specific pre-trained BERT models for relation extraction? [article]

Anfu Tang, Claire Nédellec
2021 arXiv   pre-print
The DrugProt track at BioCreative VII provides a manually-annotated corpus for the purpose of the development and evaluation of relation extraction systems, in which interactions between chemicals and  ...  Recently many studies have been conducted on the topic of relation extraction.  ...  Therefore, to achieve this goal, we propose a novel BERT-based neural network that integrates constituent information, to distinguish the novel model from the vanilla  ... 
arXiv:2112.02955v1 fatcat:xjdbtgfr2fg77fkcfi7m2gymea

Using Recurrent Neural Networks for Part-of-Speech Tagging and Subject and Predicate Classification in a Sentence

David Muñoz-Valero, Luis Rodriguez-Benitez, Luis Jimenez-Linares, Juan Moreno-Garcia
2020 International Journal of Computational Intelligence Systems  
Two different deep neural networks are used to complete this process.  ...  Finally, a comparison of the results obtained for each individual network with similar tools such as NLTK, pyStatParser and spaCy is made.  ...  For this reason, we have compared with the labelling tool provided by NLTK for the first neural network, and with a parser called pyStatParser for the second neural network.  ... 
doi:10.2991/ijcis.d.200527.005 fatcat:a2epghhqyrdjzli3jlxvrjpvei
« Previous Showing results 1 — 15 out of 467 results