Optimal Web Page Classification Technique Based on Informative Content Extraction and FA-NBC
A. M. James Raj, Flory Francis, P. Julian Benadit · 2016
‘Web Mining’ refers to a group of techniques that derive interesting patterns of information from the World Wide Web. ‘Web content mining’, which is one of the Web mining techniques aims at retrieving interesting patterns of information from the raw data that exist in the Web pages. The source data primarily contain textual data in Web pages such as the words and their tags. General applications on these data are content-based categorization and content-based ranking. This paper proposes a method that is made up of three phases for classifying the Web pages, viz., feature extraction, information learning and classification. It first extracts object based features and utilizes these features to retrieve informative contents. Next it takes both terms and HTML tags at the same time on a Web page as features to extract the informative contents from the Web pages. The decision tree learning method is used to extract the rules from the features calculated. Based on the rules extracted, the Web pages are classified using optimal Firefly Algorithm (FA) based Naive Bayes Classifier (FA-NBC) in the final phase. Here the FA are used to optimize the rules extracted. The method was implemented in Java and its performance is compared with the existing classifiers. It is shown that this new method provides better performance than the existing classifier KNN.