A Web Page Classification Technique with Textual Content Analysis Using NN-PCA for Objectionable Web Page Classification

As the Internet has recently been rapidly expanded; we can find information easily and quickly. A lot of useful information exists on the internet, but there is also harmful information involving pornography, adult content, is not appropriate for all users. This is particularly problematic when children are able to access the objectionable material with ease. Pornography web content is one of the biggest harmful resources that pollute the healthy mind of children and teenagers. Several content based analysis approaches had been proposed to avoiding objectionable and other offensive material accessed by the children. Internet users have begun to protect themselves and their wards by using so called web content filters, which allow access to legitimate content and disallow access to objectionable, illegal, pornographic, and other problematic content. This paper, proposes a new content based classification scheme for objectionable web documents filtering. Proposed method uses a neural network with input obtained by Principal Component Analysis (PCA). Each web page is represented by the term weighting scheme. As the number of unique words in the collection set is big, the PCA has been used to select the most relevant features for the classification. These feature vectors are then used as the input to the neural network for classification. We have conducted results by taking four data sets containing different web pages to test the performance of classifier in various scenarios. The experimental evaluation demonstrates that the proposed method provides outstanding classification accuracy for objectionable document classification