Adult content filtering through compression-based text classification

Abstract

Internet is a powerful source of information. However, some of the information that is available in the Internet, cannot be shown to every type of public. For instance, pornography is not desirable to be shown to children. To this end, several algorithms for text filtering have been proposed that employ a Vector Space Model representation of the webpages. Nevertheless, these type of filters can be surpassed using different attacks. In this paper, we present the first adult content filtering tool that employs compression algorithms to represent data that is resilient to these attacks. We show that this approach enhances the results of classic VSM models. © 2013 Springer-Verlag Berlin Heidelberg.