Semantic approach for identifying harmful sites using the link relations

Abstract

Technologies based on the internet are improving continuously and so are the harmful websites such as pornography or illegal gambling websites. In addition, it is the characteristics of websites that the changes made to the web address or its contents take effect almost instantaneously. Therefore, it is not easy to identify harmful websites from those that are not. There are two ways to make such decision: manual examination and automated searching for certain texts, videos, or sounds. These two methods require a lot of time. In this paper we propose a method for identifying harmful websites by analyzing the relationship between websites instead of the contents. © 2014 IEEE.