Grid-enabled Automatic Web Page Classification

There are billions of Web pages on the World Wide Web and the number continues to grow exponentially. Much research has been conducted on the efficient retrieval and classification of Web-based information. One of the big challenges those approaches face is the performance issue. It may take a long time for an algorithm to return a result across the large set of data that is typical in accessing the web. This paper describes a grid-enabled approach for automatic Web page classification that applies the vector space model information retrieval strategy. The approach can efficiently retrieve Web pages across a number of Web site groupings (such as those provided by Web domains) and classify them in an organized manner. A prototype is implemented and initial empirical studies are conducted to demonstrate feasibility. The contributions of this paper are: (1) Application of grid computing to improve performance of automatic Web page classification; (2) Enhanced classification results by exploring the parameters of automated relevance feedback.