Grammatical Interference: Learning Syntax from Sentences : Third International Colloquium, ICGI-96 Montpellier, France, September 25-27, 1996 Proceedings

"This book constitutes the refereed proceedings of the Third International Colloquium on Grammatical Inference, ICGI-96, held in Montpellier, France, in September 1996. The 25 revised full papers contained in the book together with two invited key papers by Magerman and Knuutila were carefully selected for presentation at the conference. The papers are organized in sections on algebraic methods and algorithms, natural language and pattern recognition, inference and stochastic models, incremental methods and inductive logic programming, and operational issues."--PUBLISHER'S WEBSITE Learning grammatical structure using statistical decision-trees....Pages 1-21 Inductive inference from positive data: from heuristic to characterizing methods....Pages 22-47 Unions of identifiable families of languages....Pages 48-58 Characteristic sets for polynomial grammatical inference....Pages 59-71 Query learning of subsequential transducers....Pages 72-83 Lexical categorization: Fitting template grammars by incremental MDL optimization....Pages 84-94 Selection criteria for word trigger pairs in language modeling....Pages 95-106 Clustering of sequences using a minimum grammar complexity criterion....Pages 107-116 A note on grammatical inference of slender context-free languages....Pages 117-125 Learning linear grammars from structural information....Pages 126-133 Learning of context-sensitive language acceptors through regular inference and constraint induction....Pages 134-145 Inducing constraint grammars....Pages 146-155 Introducing statistical dependencies and structural constraints in variable-length sequence models....Pages 156-167 A disagreement count scheme for inference of constrained Markov networks....Pages 168-178 Using knowledge to improve N-Gram language modelling through the MGGI methodology....Pages 179-190 Discrete sequence prediction with commented Markov models....Pages 191-202 Learning k-piecewise testable languages from positive data....Pages 203-210 Learning code regular and code linear languages....Pages 211-221 Incremental regular inference....Pages 222-237 An incremental interactive algorithm for regular grammar inference....Pages 238-249 Inductive logic programming for discrete event systems....Pages 250-261 Stochastic simple recurrent neural networks....Pages 262-273 Inferring stochastic regular grammars with recurrent neural networks....Pages 274-281 Maximum mutual information and conditional maximum likelihood estimations of stochastic regular syntax-directed translation schemes....Pages 282-291 Grammatical inference using Tabu Search....Pages 292-300 Using domain information during the learning of a subsequential transducer....Pages 301-312 Identification of DFA : Data-dependent versus data-independent algorithms....Pages 313-325

This book constitutes the refereed proceedings of the Third International Colloquium on Grammatical Inference, ICGI-96, held in Montpellier, France, in September 1996.
The 25 revised full papers contained in the book together with two invited key papers by Magerman and Knuutila were carefully selected for presentation at the conference. The papers are organized in sections on algebraic methods and algorithms, natural language and pattern recognition, inference and stochastic models, incremental methods and inductive logic programming, and operational issues.

Parsing a natural language sentence can be viewed as making a sequence of disambiguation decisions: determining the part-of-speech of the words, choosing between possible constituent structures, and selecting labels for the constituents.