Good word attack spam filtering model based on artificial immune system

Current spam filtering algorithm based on artificial immune system considers little regarding to the good word attack, so in this paper, we put forward an immune-based good word attack spam filtering model called GWAIS, which uses the Mutual Information Difference as the evaluation function to discard the good words in the spam and the spam words in the normal email during the generation of the gene library, so that it can improve the quality of the antibody. The experimental results show that GWAIS has better classification performance than that of DTDEF (Danger Theory Based Junk Mail Dynamic Filtering System).