Title :
Spam filtering system study based on 2v-SVM
Author :
Wang, Anna ; Fan, Weihao ; Wu, Jie ; Shi, Yongyue ; Li, Dan
Author_Institution :
Sch. of Inf. Sci. & Eng., Northeastern Univ., Shenyang
Abstract :
A novel spam filtering algorithm based on 2v-SVM is proposed on the basis of researching of existing spam filtering algorithms in the paper. 2v-SVM can address the difficulties that arise when the class frequencies in training data do not accurately reflect the true prior probabilities of the classes, which is more superiority than standard SVM. Experiment results show that this method can effectively improve spam filtering accurateness.
Keywords :
e-mail filters; pattern classification; security of data; support vector machines; telecommunication security; unsolicited e-mail; 2v-SVM; class frequencies; spam filtering system; training data; Filtering algorithms; Frequency; Information filtering; Information filters; Internet; Support vector machine classification; Support vector machines; Switches; Training data; Unsolicited electronic mail; 2v-SVM; classification; feature vector; spam filtering;
Conference_Titel :
Intelligent Control and Automation, 2008. WCICA 2008. 7th World Congress on
Conference_Location :
Chongqing
Print_ISBN :
978-1-4244-2113-8
Electronic_ISBN :
978-1-4244-2114-5
DOI :
10.1109/WCICA.2008.4594517
