A COLLABORATIVE INTERNET DOCUMENTS ACCESS SCHEME USING ACIRD

Abstract

In this paper, we present a collaborative intelligent Internet multi-web sites documents search system using ACIRD. ACIRD is a system that automatically learns the classification knowledge from Web pages and applies the knowledge to automatic classification of Web pages to some classes in a class hierarchy. Data mining technique is used to learn the association of terms to discover the hidden semantic connections between terms. With the capabilities of ACIRD, it is straigh tforward to extend ACIRD to collaborate multi-web site document access. Based on the learned classification knowledge, a collaborative two-phase search engine is proposed, which dispatches queries to distributed Web sites to match documents and presents hierarchically navigable results to the Internet users rather than conventional ranked flat results.