Spam detection with genetic optimized artificial immune system

Abstract

Spam has become one of the most serious universal problems, which causes problems for almost all computer users. These problems such as lost productivity, wasting user’s time and occupying network bandwidth, causes a big problem for companies and organizations. This study presents a hybrid machine learning approach inspired by the Artificial Immune System (AIS), and Genetic algorithm for effectively detect the Spams. The Clonal Selection Algorithm (CLONALG) is one of the famous implementations of the AIS, which is inspired by the clonal selection theory of acquired immunity, which has shown success on broad range of engineering problem domains. This algorithm is quietly similar to Genetic Algorithm in terms of architecture and behavior. In this study, Comparisons are drawn with AIS and GA-AIS classifiers and it is shown that the proposed system performs better results than the original AIS.