LAUSR

The resources in the World Wide Web are rising to large extent. Furthermore, the services and applications provided by the web are directly proportional to its growth. Hence, web traffic is large, and gaining access to these resources incurs user-perceived latency. Although the latency can never be avoided, it can be minimized largely. Web prefetching is identified as a technique to minimize this latency wherein it anticipates user’s future requests and fetches them into the cache prior to an explicit request made. Since web objects dispersed across the web are of various types, a new algorithm is being proposed that concentrates on prefetching embedded objects including audio and video files. Further, clustering is employed using Adaptive Resonance Theory2 neural network so as to prefetch embedded objects as clusters. For comparative study, the web objects are clustered using state-of-the-art clustering techniques and Adaptive Resonance Theory1. The clustering results confirm the supremacy of the adaptive ART2, and thereby prefetching web objects in clusters is observed to produce higher hit rate than all other techniques.