Paper: A Discriminative Approach to Japanese Abbreviation Extraction

ACL ID I08-2127
Title A Discriminative Approach to Japanese Abbreviation Extraction
Venue International Joint Conference on Natural Language Processing
Session Main Conference
Year 2008

This paper addresses the difficulties in rec- ognizing Japanese abbreviations through the use of previous approaches, examining ac- tual usages of parenthetical expressions in newspaper articles. In order to bridge the gap between Japanese abbreviations and their full forms, we present a discrimina- tive approach to abbreviation recognition. More specifically, we formalize the abbrevi- ation recognition task as a binary classifica- tion problem in which a classifier determines a positive (abbreviation) or negative (non- abbreviation) class, given a candidate of ab- breviation definition. The proposed method achieved 95.7% accuracy, 90.0% precision, and 87.6% recall on the evaluation corpus containing 7,887 (1,430 abbreviations and 6,457 non-abbreviation) instances of paren- thetical expres...