Presentation 2003/7/24
Agent-based Web Pages Classify Model A Experimental Report Part 1
Hirokazu ARATANl, Shigeru FUJITA, Kenji SUGAWARA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent year, crawler-based search engines such as Google have been popular for infomation retreival.However, there is the topic drift problem which is caused by such web pages as the same words with search keywordsor many hyperlinks are described repeatedly. As results of this problem, web pages which dont include useful contentenough for searchers appear in retrieval results at high rank. In this paper, we provide the solution of the topicdrift problem. And we have evaluated our retrieval model in experimental data set.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Web page classify system / Multi-agent system / Topic drift problem
Paper # AI2003-18
Date of Issue

Conference Information
Committee AI
Conference Date 2003/7/24(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 Artificial Intelligence and Knowledge-Based Processing (AI)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Agent-based Web Pages Classify Model A Experimental Report Part 1
Sub Title (in English)
Keyword(1) Web page classify system
Keyword(2) Multi-agent system
Keyword(3) Topic drift problem
1st Author's Name Hirokazu ARATANl
1st Author's Affiliation Graduate School, Chiba Institute of Technology()
2nd Author's Name Shigeru FUJITA
2nd Author's Affiliation Department of Computer Science, Faculty of Computer and Infomation Science, Chiba Intitute of Technology
3rd Author's Name Kenji SUGAWARA
3rd Author's Affiliation Department of Network Science, Faculty of Computer and Infomation Science, Chiba Intitute of Technology
Date 2003/7/24
Paper # AI2003-18
Volume (vol) vol.103
Number (no) 243
Page pp.pp.-
#Pages 6
Date of Issue