Let The Anti-spam Technology Research And Prototype Implementation

The topic of this paper is "the research and prototype implementation of the techniquerelated to anti_spam". It proposes and researches the content filtering technique and thearchitecture of the anti_spam system.The technique of the e_mail filtering has important value for application. Accordingrelevant department statistic, before the video application has not become the main businessthe e_mail is above 70 percent in internet application, but above 10 percent is spam in 1.4billion e_mail every day. So the spam occupies a lot of resoure of network, system andmemory.Firstly, the paper introduces the e_mail protocols, anti_spam technique, and Karhunetn-Loeve algorithm, genitic algorithm, bayes algorithm. Secondly, based on above algorithms,they are used for the anti_spam system by us. The paper emphasises on the chinese wordspliting, feature extracting in e_mail content text, and then the system reduces the dim-ensionwith k-l algorithym and selects sub_set of feature through the genitic algorithm. Now theusing methods decline the complexity of the e_mail dealing process. It enhances the self_adapting ablity of the system that the bayes studies and the decision of bayes risk. Therefore,we got the high accuracy for e_mail classifying. Thirdly, the paper analyses and discusses thedevelopment for the anti_spam system arranged according to the order of the analysing, desi-gning, implementing and testing, and describes the design and implementation of the basicalsystem and console. At last, we attempt to explore the harmful image filtering methods in thispaper.