A Novel ART1NN Clustering and Pre-fetching Technique for Reducing Web Latency

The World Wide Web is a huge information repository. It is being used for e-business, e-education, ecommerce, personalization, web site design and management, network traffic analysis and to predict user´s actions. Its popularity is resulting in heavy traffic in the Internet which in turn causes the users to experience a long latency when retrieving the Web pages. Potential sources of latency are the Web servers´ heavy load, network congestion, low bandwidth, bandwidth underutilization, and propagation delay. In order to reduce the latency, this paper proposes a novel approach that makes use of ART1NN Clustering and Prefetching technique, to fetch the pages for the user community before request them.