Detecting link hijacking by web spammers

Abstract

Since current search engines employ link-based ranking algorithms as an important tool to decide a ranking of sites, Web spammers are making a significant effort to manipulate the link structure of the Web, so called, link spamming. Link hijacking is an indispensable technique for link spamming to bring ranking scores from normal sites to target spam sites. In this paper, we propose a link analysis technique for finding link hijacked sites using modified PageRank algorithms. We performed experiments on the large scale Japanese Web archive and evaluated the accuracy of our method. Detection precision of our approach was improved about 25% from a naive approach. © Springer-Verlag Berlin Heidelberg 2009.