Bayes Data Mining Technique And Its Application In Anti-Spam

With the rapid development of the Internet, Electronic mail is becoming a kind of the fastest and the most economical means of communication. As a tool for information exchange, E-mail also serves as a kind of business advertisement means. User receives all kinds of advertisement mails on the Internet while he receives useful information. As a result, user spends much time on dealing with these so-called "SPAM". How to deal with E-mail, retaining useful mails and filtering SPAM, is a big problem of users' concern. This is the so-called "Anti-Spam". Although some current systems have adopted some techniques to proceed Anti-Spam, these techniques have some shortages. Accordingly, it is significant to explore an effective Anti-Spam system.Classification is one of the most important techniques in data mining. It analyses and learns a large quantity of relevant data and establish classification model of corresponding issue space to predict future data. Classify method contains Decision Tree, Neural Network and Bayesian Classifier etc. Bayesian Classifier is derived from statistics and is widely used in the area of classification for the simple method can classify texts quickly with high accuracy.Based on the principle of Bayesian Classifier, the design and implement of the Anti-Spam model is proposed in this paper. An algorithm based on Genetic Algorithm to optimize the model is suggested too.We also explore the threshold in the Anti-Spam model based on Bayesian Classifier.The experimental results show that these algorithms have fairly satisfactory performance.