论文信息 - Text Corpora, Local Grammars and Prediction

Text Corpora, Local Grammars and Prediction

We present a corpus-based method for identifying and learning patterns describing events in a specific domain by examining the manner in which: (a) a small number of keywords in the domain are distributed throughout the corpus; and, (b) a local grammar that is idiosyncratic of a class of events in the domain, governs the usage of the keywords. We tested our method against a corpus of 3.63 million words. The results show promise. More importantly, the method can be applied to any arbitrary domains.

[1] M. Gross. Local grammars and their representation by finite automata , 1992 .

[2] M. Gross. The Construction of Local Grammars , 1997 .

[3] John Sinclair,et al. Corpus, Concordance, Collocation , 1991 .

[4] G Demetriou,et al. Two applications of information extraction to biological science journal articles: enzyme interactions and protein structures. , 1999, Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing.

[5] Jorge Baptista,et al. A Computational Lexicon of Portuguese for Automatic Text Parsing , 1999 .

[6] Z. Harris. A Theory of Language and Information: A Mathematical Approach , 1991 .

[7] Kristin Davidse. Gill Francis, Susan Hunston and Elizabeth Manning (eds.) Collins COBUILD Grammar Patterns 1: Verbs , 1998 .

[8] Oliver Mason,et al. Automatic Processing of Local Grammar Patterns , 2004 .

[9] Key-Sun Choi,et al. A Local Grammar-based Approach to Recognizing of Proper Names in Korean Texts , 1997, VLC.

[10] Ralph Grishman,et al. Customization of information extraction systems , 1997 .

[11] Frank Smadja,et al. Retrieving Collocations from Text: Xtract , 1993, CL.