A spam filtering model based on immune mechanism

With the development of network, some mail business growing has become a pressing problem in the internet. The problem for the traditional method of spam filtering can not effectively identify the unknown and variation characteristics, artificial immune system exists diversity, immune memory, adaptive and self learning ability, adopt the idea of to mail filtering, and design an improved spam filtering model based on immune mechanism. The model describes the concepts of self, nonself, antibody, vaccine and antigen, and introduced the process of evolution of detector and antigen presentation, make the antibody in all kinds of detectors use vaccines as a medium to communicate with each other, share antibody, increases flexibility of detectors, and effectively extract information and variability of spam. Using CCERT mail dataset of the model was trained and tested, the model and other models by the comparative test results show that the proposed model has better performance, and effectively improve the spam precision, recall characteristics.