Machine Learning Techniques in Web Content Mining: A Comparative Analysis

Abstract

With incessantly growing amount of information published over Web pages, the World Wide Web (WWW) has become prolific in the field of data mining research. The heterogeneous and semi-structured nature of Web data has made the process of automated discovery a challenging issue. Web Content Mining (WCM) essentially uses data mining techniques to effectively discover knowledge from Web page contents. The intent of this study is to provide a comparative analysis of Machine Learning (ML) techniques available in the literature for WCM. For analysis, the article focuses on issues such as representation techniques, learning methods, datasets used and performance of each method as a criterion. The survey observes that some of the traditional ML algorithms have been efficiently used to work on Web data. Finally, the paper concludes citing some promising issues for further research in this domain.

Classification of Web Documents Using Hybrid Feature Selection.

“…Anami, Basavaraj S., Ramesh S. Wadawadagi, and Veerappa B. Pagi. [16] used Self Organizing Maps for the text and document clustering using Bag of Words with n-grams.…”

Web Page Content Block Identification with Extended Block Properties

“…However, current technologies are increasingly interconnected, so the content must be designed for machines to read, rather than just humans [1]. Machine-adapted labeling of content on web pages is a base for automated content extraction, data mining, content transformation, and other needs [2]. However, the existing HTML standard is slowly moving away from presentation over data.…”