Extraction and Classification of Term Definitions Using Named Entity Extraction from News Articles
-
- KOYAMA Makoto
- Knowledge Media Laboratory, Corporate Research & Development Center, TOSHIBA CORPORATION
-
- SAKAI Tetsuya
- Knowledge Media Laboratory, Corporate Research & Development Center, TOSHIBA CORPORATION
-
- Manabe Toshihiko
- Knowledge Media Laboratory, Corporate Research & Development Center, TOSHIBA CORPORATION
Bibliographic Information
- Other Title
-
- 新聞記事からの用語定義の抽出と固有表現クラスに基ずく分類
Search this article
Abstract
In this paper, we propose a system that uses Japanese newspaper corpora for extracting and classifying term definitions to expand the knowledge of a natural language system such as a question answering system. The system classifies term definitions based on semantic classes obtained through named entity extraction and words obtained through morphological analysis. In an experiment using news articles, the system classifies term definitions by 14 semantic classes and achieves 82.1% precision and 50.8% recall.
Journal
-
- IPSJ SIG Notes
-
IPSJ SIG Notes 2004 45-51, 2004
Information Processing Society of Japan (IPSJ)
- Tweet
Details 詳細情報について
-
- CRID
- 1573950401828079232
-
- NII Article ID
- 110002949406
-
- NII Book ID
- AN10115061
-
- Text Lang
- ja
-
- Data Source
-
- CiNii Articles