Presentation 2007-05-25
A detecting method of BBS vandalism based on correlation of comments
Yu ICHIFUJI, Susumu KONNO, Hideaki SONE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Electronic bulletin board system (BBS) has problems of vandalism. It is necessary for an operator to find such problems quickly when happened. For detecting such vandalism in BBS, we propose three methods to evaluate the probability based on bayesan inference. One is evaluation of the probability simply. Second employs combination of comments, and calculates its probability. Third introduces arrangement of comments as context. The result of evaluation of three methods are shown, and efficiency to find such vandalism is discussed.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) BBS / vandalism / bayesian inference
Paper # SITE2007-4
Date of Issue

Conference Information
Committee SITE
Conference Date 2007/5/18(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 Social Implications of Technology and Information Ethics (SITE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A detecting method of BBS vandalism based on correlation of comments
Sub Title (in English)
Keyword(1) BBS
Keyword(2) vandalism
Keyword(3) bayesian inference
1st Author's Name Yu ICHIFUJI
1st Author's Affiliation Graduate School of Information Sciences, Tohoku University()
2nd Author's Name Susumu KONNO
2nd Author's Affiliation Graduate School of Information Sciences, Tohoku University
3rd Author's Name Hideaki SONE
3rd Author's Affiliation Information Synergy Center, Tohoku University
Date 2007-05-25
Paper # SITE2007-4
Volume (vol) vol.107
Number (no) 66
Page pp.pp.-
#Pages 5
Date of Issue