Enriching the knowledge sources used in a maximum entropy part-of-speech tagger

被引：415

作者：

Toutanova, K ^{[1
]}

Manning, CD ^{[1
]}

机构：

[1] Stanford Univ, Dept Comp Sci, Stanford, CA 94305 USA

来源：

PROCEEDINGS OF THE 2000 JOINT SIGDAT CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING AND VERY LARGE CORPORA | 2000年

关键词：

D O I：

10.3115/1117794.1117802

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents results for a maximum-entropy-based part of speech tagger, which achieves superior performance principally by enriching the information sources used for tagging. In particular, we get improved results by incorporating these features: (i) more extensive treatment of capitalization for unknown words; (ii) features for the disambiguation of the tense forms of verbs; (iii) features for disambiguating particles from prepositions and adverbs. The best resulting accuracy for the tagger on the Penn Treebank is 96.86% overall, and 86.91% on previously unseen words.

引用

页码：63 / 70

页数：8

共 11 条

[1] BAKER CL, 1995, ENGLISH SYNTAX
[2] Berger AL, 1996, COMPUT LINGUIST, V22, P39
[3] Brants T, 2000, 6TH APPLIED NATURAL LANGUAGE PROCESSING CONFERENCE/1ST MEETING OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, PROCEEDINGS OF THE CONFERENCE AND PROCEEDINGS OF THE ANLP-NAACL 2000 STUDENT RESEARCH WORKSHOP, P224
[4] BRILL E, 1994, PROCEEDINGS OF THE TWELFTH NATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOLS 1 AND 2, P722
[5] Charniak E, 2000, 6TH APPLIED NATURAL LANGUAGE PROCESSING CONFERENCE/1ST MEETING OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, PROCEEDINGS OF THE CONFERENCE AND PROCEEDINGS OF THE ANLP-NAACL 2000 STUDENT RESEARCH WORKSHOP, pA132
[6] Jelinek F., 1997, Statistical Methods for Speech Recognition
[7] Manning C.D., 1999, FDN STAT NATURAL LAN
[8] MIKHEEV A, 1999, THESIS U EDINBURGH
[9] Ratnaparkhi A., 1998, MAXIMUM ENTROPY MODE
[10] Ratnaparki A, 1996, P C EMP METH NAT LAN, P133

← 1 2 →