Reach Junk Mail Intelligent Filter System Based On Content

The E-mail is the most commonly used on the Internet, also the important tool which people carries on communion mutually. But the e-mail has initiated the massive questions while brings the huge convenience to people, massive e-mail is useless even harmful in network dissemination, not only consumed the massive networks resources, moreover seriously threatens the mail system user information security. More serious problem is some harmful content includes the pornographic and reactionary, not only hinders the user to use the e-mail, also endangered the national security. Therefore the effective preventing and controlling junk e-mail is realistic question of vital meaning.At the present time, the multimedia is the development direction of the E-mail system, but the text e-mail is the way to sent messages and transfer documents for most consumers in common use. Because of the difference of the interests and differentiation ability between customers, the filtration method based on the e-mail content will be a very effective way to realize the individuation e-mail filtration by using Bayesian algorithm.The machine learning technology provided the method to filter the e-mail. This paper emphasized introduces with how uses the Bayesian algorithm to filter the e-mail. This paper first explains the harm of current junk e-mail and the importance and necessity of developing counter junk e-mail, analyzes the current some mainstreams counter junk e-mail system principles, proposes junk e-mail filtered system based on the content is the current development mainstream direction.This paper carries on analysis to e-mail system related agreement, introduces the POP agreement, SMTP agreement as well as the way of e-mail text MIME packs in detail, it has the extremely vital meaning for studying the e-mail filter system working principle.Then, this paper introduced the survey of Bayesian sorter, general model, principle and characteristic, and uses Bayesian algorithm to expatiate on e-mail filter method. Although the e-mail system is turning towards the multimedia direction to develop, the overwhelming majority user still used the text e-mail to transmit the information and document. Therefore, the method using Bayesian algorithm based on e-mail content is one of extremely effective methods.Finally, the application above proposed Bayesian algorithm, usesprogramming technology, develops mail filter system based on the windows platform and Visual Basic programming language, and introduce: overall structure of the system and each module detailed realization step.