Title :
Research on illegal E-mails recognition based on VSM and Statistical Decision Tree
Author :
Wang, Ke-jian ; Han, Xian-zhong ; Guo, Tao
Author_Institution :
Sch. of Inf. Sci. & Technol., Agric. Univ. of Hebei, Baoding
Abstract :
This paper introduces an algorithm based on VSM algorithm and statistical decision tree (SDT) to recognize illegal e-mails. The vector space model is simple and easy to operate. At first, the vector space model (VSM ) can filter some specific words which are often used in illegal e-mails. Then, SDT can judge illegal e-mails by Semanteme analyze. After the two steps, the illegal e-mails can also be easily identified and the recognition rate of illegal E-mails has been improved by basic experiments. Theoretical analysis and basic experiments shows that the illegal emails can be recognized effectively with VSM and SDT algorithm.
Keywords :
decision trees; information filtering; statistical analysis; unsolicited e-mail; illegal e-mail recognition; information filtering; semanteme analyze; statistical decision tree; vector space model; Algorithm design and analysis; Decision trees; Electronic mail; Filters; Internet; Pattern recognition; Postal services; Space technology; Unsolicited electronic mail; Wavelet analysis; Illegal E-mails; Semanteme analyze; Statistical Decision Tree; Vector Space Model;
Conference_Titel :
Wavelet Analysis and Pattern Recognition, 2008. ICWAPR '08. International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-2238-8
Electronic_ISBN :
978-1-4244-2239-5
DOI :
10.1109/ICWAPR.2008.4635828
