Fuzzy Based Spam Filtering

Abstract

Emails have proliferated in our ever-increasing communication, collaboration and
information sharing. Unfortunately, one of the main abuses lacking complete benefits of
this service is email spam (or shortly spam). Spam can easily bewilder system because
of its availability and duplication, deceiving solicitations to obtain private information.
The research community has shown an increasing interest to set up, adapt, maintain and
tune several spam filtering techniques for dealing with emails and identifying spam and
exclude it automatically without the interference of the email user. The contribution of
this paper is twofold. Firstly, to present how spam filtering methodology can be
constructed based on the concept of fuzziness mean, particularly, fuzzy c-means (FCM)
algorithm. Secondly, to show how can the performance of the proposed FCM spam
filtering approach (coined hence after as FSF) be improved. Experimental results on
corpora dataset point out the ability of the proposed FSF when compared with the known
Naïve Bayes filtering technique.