Presentation 2012-03-08
Effect of the improvement on classification performance of the filtering system based on Bayesian and decision tree methods
Hiroyuki YAMAGUCHI, Manabu SUGII, Hiroshi MATSUNO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recent increase of spam mails is a serious problem in the Internet. bsfilter is one of the most well-used spam filters in the world, which employs Bayes's theorem for the classification of regular and spam mails. Aiming at more efficient mail classification, we have been working on the development of a new mail filtering system, in which the bsfilter and a machine learning system BONSAI are combined. In this system, after classifying regular mails by bsfilter, BONSAI is used for the classification of remaining mails. For the further improvement of this system for collecting all the regular mails, we have made the following two trials; spam mail collection in the first step by bsfilter and an appropriate threshold setting of the bsfilter. Experimental results are shown to prove the efficiency of the proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) security / mail filterring / machine learning / Bayes method
Paper # MSS2011-76
Date of Issue

Conference Information
Committee MSS
Conference Date 2012/3/1(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 Mathematical Systems Science and its applications(MSS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Effect of the improvement on classification performance of the filtering system based on Bayesian and decision tree methods
Sub Title (in English)
Keyword(1) security
Keyword(2) mail filterring
Keyword(3) machine learning
Keyword(4) Bayes method
1st Author's Name Hiroyuki YAMAGUCHI
1st Author's Affiliation Graduate School of Science and Engineering, Yamaguchi University()
2nd Author's Name Manabu SUGII
2nd Author's Affiliation Media and Infomation Technology Center, Yamaguchi University
3rd Author's Name Hiroshi MATSUNO
3rd Author's Affiliation Graduate School of Science and Engineering, Yamaguchi University
Date 2012-03-08
Paper # MSS2011-76
Volume (vol) vol.111
Number (no) 453
Page pp.pp.-
#Pages 6
Date of Issue