An effective spam filter based on a combined support vector machine approach
by Mumtaz M. Al-Mukhtar; Yasmine M. Tabra
International Journal of Internet Technology and Secured Transactions (IJITST), Vol. 4, No. 1, 2012

Abstract: The volume of mass unsolicited e-mail, often known as spam, has recently increased enormously and has become a serious threat to not only internet but also to society. It is challenging to develop spam filters that can effectively eliminate the increasing volume of unwanted e-mails automatically. The present work presents a combination of support vector machine classifier for non-linear data (using an eligible kernel function) with appropriate data pre-processing as a spam filter. Data pre-processing is a vital part of text classification where the objective is to generate feature vectors usable by SVM kernels. The pre-processing steps include HTML removal, HTML replacement, de-obfuscation and stop-word-remover. The results obtained using the pre-processing level showed an improvement in the classification level. The estimated training and classification time for different document sizes indicate that the adopted method is practical and computationally efficient. Experimental results show that the approach can enhance the filtering performance effectively.

Online publication date: Sat, 09-Aug-2014

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Internet Technology and Secured Transactions (IJITST):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com