Design And Implementation Of Spam Identification System Based On Ais

Service by e-mail services, e-mail users can be simple and quick way to reach large groups of users, which also led to the motives the malicious user send information garbage for various purposes. There are many different anti-spam technologies in the anti-spam area, but most of the technical existence of fatal flaws: a high false alarm rate, low processing performance, language-dependent strong, and not fit at the gateway to use. These filtering technologies have not filtered out of the content matching technology limitations. Therefore the spam-depth study to identify the principles of behavior and methods. This technology greatly enhanced the speed of e-mail filtering, but also was a content filtering technology to avoid the inevitable problem of high false alarm rate, making spam filtering efficient and accurate. Artificial immune system have characters with powerful recognition, learning and memory capabilities, its non-autologous autologous recognition to identify junk e-mail is a good and natural solution, and its sensitive immune feedback mechanism for identifying changing spam offers a solution Road.In this thesis, there will be the brief introductions on anti-right right on the current state of technology firstly, and then on the anit-spam basic theory, principles of artificial immune biology background and artificial immune.This study focus on behavior-based spam filters, as well as the principles of artificial immune on its applications. Then it will focus on the behavioral characteristics which e-mail header keywords reflect to identify the difference between the nomal mail and the spam. Also prove that these are distinctive characteristics of spam and the role of normal mail. Finally it will come up the anti-spam technology based on combination of artificial immune theory and behavioral characteristics to identify the spam.