Абстрактный

A Data Mining Algorithm for Long Term Web Prefetching

C.Anuradha,

The World Wide Web is the Internet’s most widely used tool for information access and dissemination, but today’s users often experience long access latency due to network congestion—particularly during peak hours and big events, such as the Olympic Games. Caching frequently used data at proxies close to clients is an effective way to alleviate these problems. Specifically, caching can reduce load on both the network and servers (by localizing the traffic) and improve access latency (by satisfying user requests from local storage rather than remote servers). Cache Prefetching is related to replacement, but unlike data caching, which waits on object requests, Prefetching proactively preloads data from the server into the cache to facilitate near-future accesses. However, a cache Prefetching policy must be carefully designed: if it fails to predict a user’s future accesses, it wastes network bandwidth and cache space. The prediction mechanism thus plays an important role. We are proposing the system where predicting the user’s future access is done by analyzing the two or three pages of his access. The pages found in accordance with the analyzed pages will be shown to the user. The user can select his required pages from that list.