An Ontology-based Term Weighting Technique for Web Document Categorization

Abstract

Web document classification has become crucial as there has been a massive increase in the magnitude of web pages across the web. In the research community, an efficient approach to this problem is based on machine learning techniques. Ontology forms the heart of knowledge representation for any domain. This paper proposes an ontology-based term weighting technique which is novel and efficient for the classification of web pages. The proposed approach builds domain ontology and selects the features that significantly improve the prediction performance. Experiments were conducted on domain based web pages and classification performance was calculated with state of the art classification algorithms. The experimental analysis demonstrates that the proposed approach produces significantly better results compared to the traditional keyword-based approaches.