The Study And Application Of Chinese-Spam Filtering Technology

With the rapid development of Internet and it's application, E-mail has become one of the fastest and the most economical ways in daily communication. At the same time, the flooding of all kinds of Spam has become a headache problem for human and society. Mail system security attracted wide attentions and became a research focus in industry.After analyzing the SMTP protocol and the feature of E-mail format in Unix/Linux system, this thesis points out the bugs of SMTP protocol and introduces the methods of tracing the source of Spam. On the basis of research in the Anti-Spam technology, the author put forward a project of Chinese-Spam filtering by combining the methods on the basis of the rule and the statistics, and applied the Chinese word segmentation technology to Chinese-Spam filtering, and solved the auto update technology in mail training set and Chinese characters filtering rule by machine learning.Because of the great difference in language between English Mail and Chinese Mail, Chinese Mail has it's own characteristics in Chinese information processing. This thesis discusses the technology of mail preprocess, Chinese word segmentation and feature selection. At last, on the basis of researches mentioned above, the author designed a kind of Chinese-Spam filtering system and realized it.