A Spam Discrimination Based on Mail Header Feature and SVM
Miao Ye, Tao Tang, Mai Fan-jin, Xiaohui Cheng · 2008
The traditional anti-spam techniques like black and white list can not meet the needs of the spam filter nowadays. Some machine learning techniques become very popular in the research of spam filter. Support vector machine is one of the most excellent methods in classifying. But these techniques are usually applied to spam identity based on the mail body textual content only, seldom discussing about mail header. This paper hereby proposes the spam discrimination model based on SVM, and uses SVM to sort out mail according to the feature of mail headers. By feature abstraction carried out on the mails dataset (CDSCE) with C++ program and SVM classifying. Experimental result indicates that the proposed model can effectively improve the accuracy of spam identification.