Evaluating Arabic spam classifiers using link analysis

Abstract

The rank of pages within the search engine results page (SERP) is important especially for commercial sites. It is also important for those establishments which try to be at the top 10 of SERP to gain more visitors. To be within the top 10 results of SERP means also being visible to a larger number of audiences. On the other hand, gaining a lower rank means being less visible and eventually gaining less revenues relative to competitive pages which are ranked higher. Therefore the Web has a number of pages which adopt spamming techniques to deceive search engines and gains a higher rank than what they really deserve. Link-based technique is one of the techniques used to achieve this goal. This study aims to detect the link-based spamming techniques used within Arabic spam Web pages. The conducted tests reveal that link-based spamming technique is used significantly within Arabic spammed Web pages.