Survey on Web Spam Detection using Link and Content Based Features

Abstract

Web spam is one of the recent problems of search engines because it powerfully reduced the quality of the Web page. Web spam has an economic impact because spammers provide a large free advertising data or sites on the search engines and so an increase in the web traffic volume. In this paper we Survey on efficient spam detection techniques based on a classifier that combines new link based features with language models. Link Based features are related to qualitative data extracted from the web pages and also to the qualitative properties of the page links. Spam technique applies LM approach to different sources of information from a web page that belongs to the context of a link in order to provide high quality indicators of web spam. Specifically Detection technique applied the Kullback Leibler divergence on different combinations of these sources of information in order to characterize the relationship between two linked pages.