贝页斯垃圾邮件分类系统成本参数调整对系统精度的影响
|
摘要
为降低垃圾邮件系统分类计算的误码率,分析了贝页斯垃圾邮件过滤系统对目标邮件的自动检测过程,从系统过滤质量和用户容错两个方面研究系统成本定义.在不同样本集合及其属性空间内,对于词语还原和间断表的开启与关闭,重点分析成本参数λ,通过调整成本参数分析贝页斯过滤系统在多种假定下邮件处理结果,完善系统建模定义标准,优化应用系统建模,提高系统过滤质量.实验结果证明该解决方案是可行的.
In order to reduce the error rate in classification calculation of anti-spam filtering system, the automatic testing process of target email in the Bayesian anti-spam filtering system was analyzed, the definition of system cost was researched from two aspects of system filtering quality and user fault tolerance. Cost parameters were analyzed in the collection of different sample sets and attribute spaces, with disabling and enabling lemmatizer and stop-list. By adjusting the cost parameters, the results of the Bayesian filtering system in various assumptions were analyzed, the standard of system modeling was optimized, and system filtering quality was upgraded. The results prove that the scheme is feasible.
In order to reduce the error rate in classification calculation of anti-spam filtering system, the automatic testing process of target email in the Bayesian anti-spam filtering system was analyzed, the definition of system cost was researched from two aspects of system filtering quality and user fault tolerance. Cost parameters were analyzed in the collection of different sample sets and attribute spaces, with disabling and enabling lemmatizer and stop-list. By adjusting the cost parameters, the results of the Bayesian filtering system in various assumptions were analyzed, the standard of system modeling was optimized, and system filtering quality was upgraded. The results prove that the scheme is feasible.
引文
[1]Kush E N.Learning to remove internet advertisements[C]//Proceedings of the 3rd International Conference on Autonomous Agents.Seattle,Washington:[s.n.],1999:175-181.
[2]Hall R J.How to avoid unwanted E-mail[J].Communication of ACM,1998,41(3):88-95.
[3]Cohen W W.Learning rule that classify E-Mail[C]//Proceedings of the AAAI Spring Symposium on Machine learning in information Access.Stanford,California:[s.n.],1996:85-90.
[4]崔超,吴双,张宪忠,等.基于贝叶斯概率理论的防火墙技术研究[J].北京理工大学学报,2012,32(8):801-804.Cui Chao,Wu Shuang,Zhang Xianzhong,et al.Firewall technology based on Bayesian probability theory[J].Transactions of Beijing Institute of Technology,2012,32(8):801-804.(in Chinese)
[5]崔超.贝叶斯网络在垃圾邮件算法中的应用研究[J].哈尔滨工业大学学报,2011,43(11):145-148.Cui Chao.Bayesian application study on arithmetic for filtering junk e-mail[J].Journal of Harbin Institute of Technology,2011,43(11):145-148.(in China)
[6]Paul G.Better Bayesian filtering[OL].[2003-01-20].http://paulgraham.com/better.html.
[7]Dan L V,SUN Jianfeng,Li Qi,et al.Model-based recognition of 3D articulated target using ladar range data[J].Applied Optics,2015,54(17): 5382-5391.
[8]Jiang Shufeng.Study on multi-strategy analysis and application of A-r algorithm[J].Information Technology Journal,2015,12(21): 6096-6097.
[2]Hall R J.How to avoid unwanted E-mail[J].Communication of ACM,1998,41(3):88-95.
[3]Cohen W W.Learning rule that classify E-Mail[C]//Proceedings of the AAAI Spring Symposium on Machine learning in information Access.Stanford,California:[s.n.],1996:85-90.
[4]崔超,吴双,张宪忠,等.基于贝叶斯概率理论的防火墙技术研究[J].北京理工大学学报,2012,32(8):801-804.Cui Chao,Wu Shuang,Zhang Xianzhong,et al.Firewall technology based on Bayesian probability theory[J].Transactions of Beijing Institute of Technology,2012,32(8):801-804.(in Chinese)
[5]崔超.贝叶斯网络在垃圾邮件算法中的应用研究[J].哈尔滨工业大学学报,2011,43(11):145-148.Cui Chao.Bayesian application study on arithmetic for filtering junk e-mail[J].Journal of Harbin Institute of Technology,2011,43(11):145-148.(in China)
[6]Paul G.Better Bayesian filtering[OL].[2003-01-20].http://paulgraham.com/better.html.
[7]Dan L V,SUN Jianfeng,Li Qi,et al.Model-based recognition of 3D articulated target using ladar range data[J].Applied Optics,2015,54(17): 5382-5391.
[8]Jiang Shufeng.Study on multi-strategy analysis and application of A-r algorithm[J].Information Technology Journal,2015,12(21): 6096-6097.