Filters








420 Hits in 4.0 sec

Connectionist natural language parsing

Dominic Palmer-Brown, Jonathan A. Tepper, Heather M. Powell
2002 Trends in Cognitive Sciences  
Connectionist parsers are assessed according to their ability to automatically learn from examples to represent syntactic structures, without being presented with symbolic grammar rules.  ...  Teaser: The extent to which connectionist systems have succeeded in parsing a wide range of realistic sentences, containing syntactic structures that are commonly found in natural language interaction,  ...  Pearlmutter et al [50] proposed an FF-MLP-based model directly inspired by the constraint-based framework presented by MacDonald et al [40, 43] Realism Beyond Context-free Grammar Modular connectionist  ... 
doi:10.1016/s1364-6613(02)01980-0 pmid:12413578 fatcat:74j5hf54wvgd3ftvh4ei6yzjeu

Parsing Complex Sentences with Structured Connectionist Networks

Ajay N. Jain
1991 Neural Computation  
A connectionist context-free parser which is not context-free but then it is not really connectionist either. In Proc. Ninth Ann. Conf. Cog. Sci. Soc., 70-77. Elman, J. L. 1990.  ...  Finding structure in time. Cog. Sci. 14(2), 179-212. Fanty, M. 1986. Context-free parsing with connectionist networks. In AIP Con- ference Proceedings number 151, J. S. Denker, ed.  ... 
doi:10.1162/neco.1991.3.1.110 fatcat:e7jq4bfrufd7vkfvqmsqxsq7xi

HCP with PSMA: A Robust Spoken Language Parser

Monirul Hasan, Venkatesh Manian, Christel Kemke
2007 International Joint Conference on Artificial Intelligence  
The HCP is a hybrid of a connectionist and a symbolic approach, which builds a neural network dynamically based on a given context-free grammar and an input sentence.  ...  We present a parsing technique that utilizes a Hybrid Connectionist Parser (HCP) extended with a Phrase Structure Matching Algorithm (PSMA) for the syntactic processing (parsing) of spoken language.  ...  He is now affiliated as software architect with a local company. This work has been supported by the Natural Sciences and Engineering Research Council of Canada (NSERC).  ... 
dblp:conf/ijcai/HasanMK07 fatcat:oqtdvnnmnfhcfnj6scxnffvotm

Connectionist syntactic parsing using temporal variable binding

James Henderson
1994 Journal of Psycholinguistic Research  
Recent developments in connectionist architectures for symbolic computation have made it possible to investigate parsing in a connectionist network while still taking advantage of the large body of work  ...  The extensive use of partial descriptions of phrase structure trees is crucial to the ability of this model to recover the syntactic structure of sentences within the constraints imposed by the architecture  ...  Grammar formalisms based on Context Free Grammars do not have this property because expectations and iteration restrictions are speci ed in terms of a node's structural position in the grammar rule, as  ... 
doi:10.1007/bf02143945 fatcat:go43hkopjjepzfqzha2so2cqim

Generalization Performance in PARSEC - A Structured Connectionist Parsing Architecture

Ajay N. Jain
1991 Neural Information Processing Systems  
This paper presents PARSEC-a system for generating connectionist parsing networks from example parses. PARSEC is not based on formal grammar systems and is geared toward spoken language tasks.  ...  PARSEC's performance is compared to that of traditional grammar-based parsing systems. IpARSEC is a generalization of a previous connectionist parsing architecture (Jain 1991) .  ...  However, the PARSEC network in the test reported here was trained only on grammatical input and still produced a degree of noise tolerance for free.  ... 
dblp:conf/nips/Jain91 fatcat:kalznvhb3zcrrmyorthlq4bsmm

PARSEC: a structured connectionist parsing system for spoken language

A.N. Jain, A. Waibel, D.S. Touretzky
1992 [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing  
We present PARSEC-a system for generating connectionist parsing networks from example parses. PARSEC is not based on formal grammar systems and has been geared towards spoken lanpage tasks.  ...  PARSEC networks exhibit three strengths important for application to speech processing: 1) they leam to parse, and generalize well compared to hand-coded grammars; 2) they tolerate several types of noise  ...  It is a system for generating connectionist parsing networks from training examples.  ... 
doi:10.1109/icassp.1992.225936 dblp:conf/icassp/JainWT92 fatcat:ecxqlrh3o5ghppay7a2l5n76fi

A corpus-based connectionist architecture for large-scale natural language parsing

Jonathan A Tepper, Heather M Powell, Dominic Palmer-Brown
2002 Connection science  
It is a modular system that learns to annotate natural language texts with syntactic structure.  ...  (For a test set of 4,128 words, precision and recall of 75% and 69% respectively were achieved).  ...  [Insert figure 7 about here] Comparisons with Previous Work As outlined in the introduction, most work on connectionist parsing has been based on language generated from artificial context-free grammars  ... 
doi:10.1080/09540090210162074 fatcat:a6lzxnoqjbcdfppgcgha6dsgri

Recent Advances in JANUS: A Speech Translation System [chapter]

A. Waibel, M. Woszczyna
1995 Speech Recognition and Coding  
Speech ia Source r~n~,u~e N-best Search LR-Parser ] I NN-Parser ] l Semantic Pra.  ...  It has initially been built based on a speech database of i 2 read dialogs, encompassing a vocabulary of around 500 words.  ...  A grammar for GENK1T is written in the same formalism as the Generalized LR Parser: phrase structure rules augmented with pseudo unification equations.  ... 
doi:10.1007/978-3-642-57745-1_66 fatcat:2w26rgf4yzcrzg4iaqzw3c37eu

SCREEN: Learning a Flat Syntactic and Semantic Spoken Language Analysis Using Artificial Neural Networks

S. Wermter, V. Weber
1997 The Journal of Artificial Intelligence Research  
Rather than using a deeply structured symbolic analysis, we use a flat connectionist analysis.  ...  While there has been some progress using statistical or connectionist language models, many current spoken- language systems still use a relatively brittle, hand-coded symbolic grammar or symbolic semantic  ...  While screen uses primarily a at connectionist language analysis, trains uses a chart parser with a generalized phrase structure grammar.  ... 
doi:10.1613/jair.282 fatcat:rogcszyhtjfl5p3h4qrtb3zjb4

SCREEN: Learning a Flat Syntactic and Semantic Spoken Language Analysis Using Artificial Neural Networks [article]

Stefan Wermter, Volker Weber
1997 arXiv   pre-print
Rather than using a deeply structured symbolic analysis, we use a flat connectionist analysis.  ...  In particular, we show how the fault-tolerance and learning capability of connectionist networks can support a flat analysis for providing more robust spoken-language processing within an overall hybrid  ...  While screen uses primarily a at connectionist language analysis, trains uses a chart parser with a generalized phrase structure grammar.  ... 
arXiv:cmp-lg/9702001v1 fatcat:eiamo7gybrhoxg2gfczpbydwte

Learning a scanning understanding for "real-world" library categorization

Stefan Wermter
1992 Proceedings of the third conference on Applied natural language processing -  
insignificant words are used for learning the classification in connectionist recurrent plausibility networks.  ...  This paper describes, compares, and evaluates three different approaches for learning a semantic classification of library titles: 1) syntactically condensed titles, 2) complete titles, and 3) titles without  ...  The symbolic chart parser built a syntactic structure for a title using a context-free grammar and a syntactic lexicon.  ... 
doi:10.3115/974499.974552 dblp:conf/anlp/Wermter92 fatcat:3og6h7ue7be23dxc7xnowdydvi

Connectionist sentence processing in perspective

M Steedman
1999 Cognitive Science  
devices based on a distinctively connectionist basis for grounded symbolist conceptual structure.  ...  as such, while auto-associative memory models of the kind pioneered by Longuet-Higgins, Willshaw, Pollack and others may be useful for grammar induction from a network-based conceptual structure as well  ...  Acknowledgments: Thanks to Gerry Altmann, Nick Chater, Morten Christiansen, Michael Collins, Geoff Hinton, Mark Liberman, Jane Neumann, and the referees for comments and advice.  ... 
doi:10.1016/s0364-0213(99)00018-x fatcat:po6ncamfovaknnl7rhpha4ax3u

Connectionist Sentence Processing in Perspective

Mark Steedman
1999 Cognitive Science  
devices based on a distinctively connectionist basis for grounded symbolist conceptual structure.  ...  as such, while auto-associative memory models of the kind pioneered by Longuet-Higgins, Willshaw, Pollack and others may be useful for grammar induction from a network-based conceptual structure as well  ...  Acknowledgments: Thanks to Gerry Altmann, Nick Chater, Morten Christiansen, Michael Collins, Geoff Hinton, Mark Liberman, Jane Neumann, and the referees for comments and advice.  ... 
doi:10.1207/s15516709cog2304_10 fatcat:kl4zwpc2jvd6hfrvhrawzhnluq

Connectionism and Determinism in a Syntactic Parser [chapter]

Stan C. Kwasny, Kanaan A. Faisal
1992 Connectionist Natural Language Processing  
This report details a set of experiments which support the claim that natural language can be syntactically processed in a robust manner using a connectionist deterministic parser.  ...  The model is trained from patterns derived from a deterministic grammar and tested with grammatical, ungrammatical and lexically ambiguous sentences. 860-2175; Bimet : FACP012Qa SAUPMOO.  ...  A preprocessing step for noun phrases is assumed , so that NPs have structure before entering the buffer.  ... 
doi:10.1007/978-94-011-2624-3_7 fatcat:ryh5qs5jxzffnbb4n5tyrop6dm

Connectionism and Determinism in a Syntactic Parser

STAN C. KWASNY, KANAAN A. FAISAL
1990 Connection science  
This report details a set of experiments which support the claim that natural language can be syntactically processed in a robust manner using a connectionist deterministic parser.  ...  The model is trained from patterns derived from a deterministic grammar and tested with grammatical, ungrammatical and lexically ambiguous sentences. 860-2175; Bimet : FACP012Qa SAUPMOO.  ...  A preprocessing step for noun phrases is assumed , so that NPs have structure before entering the buffer.  ... 
doi:10.1080/09540099008915663 fatcat:rxp4hrt3qnhvzbyfqg6fvv6zhq
« Previous Showing results 1 — 15 out of 420 results