Email Security, Filtering And Inspection Techniques Studied

With the development of Internet technology, e-mail has been used widely. Spam is very serious in China, because the e-mail filtering technology developed later in China relative to developed country. Anti-spam technology commonly used including white and black list technology; based on the keywords match the content scanning; based on the HASH of spam filtering technology; content-based spam filtering technology and other Common spam filtering technology, in which content-based spam filtering technology has been widely used.This paper analyzing all kinds of filter technology and algorithm, comprehensive use IP-based filtering, the letter head filtering,abstract filtering technology and others, design a spam e-mail filter system based on vector space model and Bayes algorithm , we analyze the disadvantages in spam e-mail filter technology by use Bayes algorithm, through made improvements to enhance the precision of filtering system.the system has self-learning ability, it can effectively filter spam through continuous training; the filter system contains: protocol agent module, pretreatment module, filter module, data module and main control module. the system is located in front of the mail server, instead of receiving e-mail, after the completion of filter,forward to the mail server.For a larger impact on system performance is how to Chinese segmentation. the system improve the efficiency by use an improved PATRICIA tree based on the segment dictionary, improve the filter Accuracy by use improved Bayes algorithm. finally, we tested the system, the system has higher efficiency and precision to filter spam.