Modeling Spammer Behavior: Naïve Bayes vs. Artificial Neural Networks

Abstract

Addressing the problem of spam emails in the Internet, this paper presents a comparative study on Naïve Bayes and Artificial Neural Networks (ANN) based modeling of spammer behavior. Keyword-based spam email filtering techniques fall short to model spammer behavior as the spammer constantly changes tactics to circumvent these filters. The evasive tactics that the spammer uses are themselves patterns that can be modeled to combat spam. It has been observed that both Naïve Bayes and ANN are best suitable for modeling spammer common patterns. Experimental results demonstrate that both of them achieve a promising detection rate of around 92%, which is considerably an improvement of performance compared to the keyword-based contemporary filtering approaches.

BSPNN: boosted subspace probabilistic neural network for email security

“…Unlike other text categorization tasks, filtering spam messages is cost sensitive (Saiful et al 2009), hence evaluation measures that account for misclassification costs. In particular, we define a cost factor λ: when a legitimate e-mail is misclassified or correctly classified, this counts as λ errors or successes, respectively.…”