Presentation 2000/5/22
A Method Decide the Importance of Word Considering Time Transition
Makoto Okada, Naotaka Nishimura, Masami Shishibori, Jun-ichi Aoe,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to evaluate a similarity between two texts, a similar text retrieval system compares words that extract from texts. However, the words in the texts change as a time transition. Therefore, this paper explains how to decide time stability of words automatically by using a decision tree algorithm (C4.5). In this paper, we define five attribute from the distribution map of word frequency. In evaluation, we used the words that extracted from reports about professional baseball of Mainichi Shinbun Newspaper (1994-1998), and it is confirmed that the defined attributes are effective for deciding time stability of words automatically.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) time stability of words / similar text retrieval / decision tree / Mainichi Shinbun Newspaper
Paper # NLC2000-1
Date of Issue

Conference Information
Committee NLC
Conference Date 2000/5/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Natural Language Understanding and Models of Communication (NLC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Method Decide the Importance of Word Considering Time Transition
Sub Title (in English)
Keyword(1) time stability of words
Keyword(2) similar text retrieval
Keyword(3) decision tree
Keyword(4) Mainichi Shinbun Newspaper
1st Author's Name Makoto Okada
1st Author's Affiliation Department of Information Science & Inteligent Systems, Faculty of Engineering, Tokushima University()
2nd Author's Name Naotaka Nishimura
2nd Author's Affiliation Department of Information Science & Inteligent Systems, Faculty of Engineering, Tokushima University
3rd Author's Name Masami Shishibori
3rd Author's Affiliation Department of Information Science & Inteligent Systems, Faculty of Engineering, Tokushima University
4th Author's Name Jun-ichi Aoe
4th Author's Affiliation Department of Information Science & Inteligent Systems, Faculty of Engineering, Tokushima University
Date 2000/5/22
Paper # NLC2000-1
Volume (vol) vol.100
Number (no) 100
Page pp.pp.-
#Pages 8
Date of Issue