Faculty Advisor

Ruiz, Carolina.

Abstract

The proliferation of spam e-mail is a rising problem. Not only an annoyance, spam wastes valuable network space and routing time. Due to high spam volume, manual filtering is inadequate. Automated spam filters exist, but better ones could help mitigate its impact. The effectiveness of Bayesian Networks for spam filtering was compared to the Naive Bayes approach, common to commercial filters. Bayesian Networks were found to produce higher quality predictions at acceptable speeds.

Publisher

Worcester Polytechnic Institute

Date Accepted

January 2005

Major

Computer Science

Project Type

Major Qualifying Project

Accessibility

Restricted-WPI community only

Advisor Department

Computer Science

Share

COinS