Spam detection using multilayer neural networks

Nyagumbirira, Brian. (2013) Spam detection using multilayer neural networks. UNSPECIFIED thesis, UNSPECIFIED.

Full text not available from this repository.
Official URL: http://hdl.handle.net/11196/1056

Abstract

Along with the popularity of the Internet technologies, especially electronic mail, there has been a growth in spam, among other problems. Several solutions have been proposed and developed to curb the problem, but email users continue to be overwhelmed by unsolicited bulk emails since spammers continue to change their spamming techniques to evade spam filters. There is an arms-race between spammers and anti-spammers, in which anti-spammers try to come up with new ways of detecting or filtering spam before it reaches the user’s inbox, and the spammers try new spamming ways to evade those solutions. Some of the ill-effects of spam include spreading of malware, cluttering of disk space and other network resources such as bandwidth, loss of employee time leading to reduced productivity, phishing and scamming. In this research, the researcher designed a spam detection algorithm using multilayer neural networks. Unlike static, hand-tuned spam filters, the perceptron learning spam detection algorithm designed and presented herein can detect new spamming variations as they occur. The researcher then experimented using a perceptron to determine whether the designed algorithm is workable, and can really be a solution.

Item Type: Thesis (UNSPECIFIED)
Uncontrolled Keywords: Web spammers,Search engine spam
Divisions: Universities > State Universities > Bindura University of Science Education
Depositing User: Mr. Edmore Sibanda
Date Deposited: 14 May 2018 09:07
Last Modified: 14 May 2018 09:07
URI: http://researchdatabase.ac.zw/id/eprint/6318

Actions (login required)

View Item View Item