A Self-learning Spam Detecting System Model Based on Memory Rules

A novel collaborative anti-spam system model based on memory rules is introduced. In this model, a self-learning spam detecting method based on neurobiology is presented. This system uses an improved chunk hashing algorithm to calculate the similarity graph of the email text. This graph, together with the arrival characteristics of email traffic, is regarded as the inputs of the spam detecting method. Finally, the feasibility of this method is validated by the simulation results.