Feature oriented fuzzy logic based web spam detection

Abstract

With Internet becoming an inevitable part of the global community and the huge amount of data resources present on the internet, the problem of malicious content present has become very serious. These spams can be introduced through many ways on the internet. Web spams involve the hike in the search engine index by manipulating the content of the web pages. To tailor the demands of the tricky spam algorithms, the hybrid use of crucial spamming features is inevitable.

This paper presents the classification of web pages as spam, undecided and non-spam using a fuzzy classifier by evaluating the spamicity, the degree to which a web page is spammed. In order to gauge the performance of the proposed scheme, we evaluated the accuracy of our classifier and also performed comparative analysis with jrip classifier and naïve Bayesian classifier through experiments.

Keywords:

• Spamicity
• Fuzzy Classification
• Keyword Density
• Semantic Correlation

Mathematics Subject Classification 2010:

• 68U99