Classification of illicit web pages using neural network

Abstract

The illicit web contents such as pornography, violence, gambling, etc, have greatly polluted the mind of web users especially children and teenagers. Due to some popular web filtering techniques like Uniform Resource Locator (URL) blocking and Platform for Internet Content Selection (PICS) checking are limited against today dynamic web content, hence content based analysis techniques with effective model are highly desired In this paper we propose textual content analysis model using entropy term weighting scheme to classify pornography and sex education web pages. We examine the entropy scheme with two other common term weighting schemes which are TFIDF and Glasgow. Those techniques are examined extensively with artificial neural network using small class dataset. In this study, we found that our proposed model archive better performance from the aspects of accuracy, convergence speed and stability.