A Study of Methods for Extracting the Boundaries of Web Spam

Abstract

As the search result ranking is getting important for attracting visitors and yielding profits, more and more people are now trying to mislead search engines in order to get higher ranking. Since link-based ranking algorithms are important tools for current search engines, web spammers are making a significant effort to manipulate the link structure of the Web, namely, link spamming. Link hijacking is one technique of link spamming. By hijacking links from normal sites to target spam sites, spammers can make search engines believe that normal sites endorse spam sites. In this research, we propose link analysis techniques for finding out link-hijacked sites using modified PageRank algorithms. We tested our methods on a large scale Japanese web archive and evaluated the accuracy.