187A COMPARISON OF MACHINE LEARNING AND NON-MACHINE LEARNING METHODS IN E-MAIL FILTERINGThe increasing volume of unsolicited bulk e-mail (spam) has generated a need for reliable anti-spam filters. Filters of this type have so far been based mostly on keyword patterns that are constructed by hand and perform poorly. In this paper an investigation is done on the performance of two approaches in the context of anti-spam filtering. The first approach uses a non-machine learning filter using black listed addresses taken from domain name system and a public spam and ham words taken from websites. The second approach uses a supervised machine learning approach based on Naïve Bayesian algorithm. The Naive Bayesian classifier has recently been suggested as an effective method to construct automatically anti-spam filters with superior performance. We investigate thoroughly the performance of the two filterson personal user E mails, and then a comparison is done on the performance of the Naive Bayesian filter to the other approach. As a result the Naïve Bayesian approach achieved accurate spam filtering, outperforming clearly the keyword-based filter of a widely used e-mail readernadiaf_1966@yahoo.comReview522015Email, anti-spam,Naïve Bayesian FilterNadia Al-BakriAssistance Lecturer in Computer Science Dept., AL Nahrain University, Baghdad, Iraq.