The Research of Spam Filtering System Based on Latent Semantic Analysis and AIS

摘要：

Along with the widespread concern of spam problem, at present, there are spam filtering system nowadays about the problem of semantic imperfection and spam filter wanting adaptation in the feature constantly change of spam. The Spam Filtering model Based on LSA and AIS is put forward. Making use of the latent semantic analysis (LSA) marks the latent feature phrase in the spam, semantic analysis is led into the Spam filtering technique, The problem of semantic imperfection is solved in spam filter; Make use of AIS on the LSA analytical foundation, email antibody database was updated from the latent feature phrase of leaks to judge spam and mistake judge spam are extended mutation. This problem of spam filtering system wanting adaptation was solved by AIS. We have designed a Spam filtering system based on this model.Our designed system was evaluated with an optional dataset. The results obtained were compared with Na(i)ve Bayes Algorithm filter experiment results show that the system based on LSA and AIS can spam filter better than Naive Bayes. The experiments show that the expected result is obtained, and the feasibility and advantage of the new spam filtering method is validated.

关键词： Latent semantic analysis AIS expand mutation Antigen Antibody Spam filtering

作者: Qiuyu Zhang Jingtao Sun Wenhan Huang Xiaowen Yan

作者单位: School of Computer and Communication,LAN Zhou University of Technology,Lanzhou, 730050, China Department of Computer science and technology,Shaanxi University of Technology,Hanzhong, 723003, Chi Shaanxi highway Group Co.Ltd,Shaanxi Xiyu highway Corporation Ltd,Hancheng,715400,China

会议类型: 国际会议

会议名称: The International Colloquium on Onformation Fusion 2007(2007年国际信息融合研讨会)

会议地点: 西安

会议语种:英文

页码: 479-483

在线出版日期: 2007-08-22（万方平台首次上网日期，不代表论文的发表时间）