To Prepare a Forecasting Proxy Server to Improve Hits Using SVM.

Abstract — Web proxy cache plays a vital role in the performance improving of World Wide Web. Also, SVM helps to wither the internet access latency and network traffic. The prediction using SVM improves the hit ratio which is beneficial for Web caching and Web prefetching environment. The research mainly focuses to apply SVM method to the problem of user action for prediction on the web. We can predict the future web page that a handler will select through replication, we determined that our approach has computable actions such as hit rate and byte hit rate of accessed page. Web proxy caching plays an important role in improving the World Wide Web routine which is actually difficult to evaluate what exactly the user’s request would be in web proxy caching practices. In this learning, we presented a new approach which depends on the expertise of SVM to learn from web proxy log data and forecast the classes of objects expected to re-visit. Hence, usage of the cache can be amplified capably. Continual experiments had revealed that support vector machine produces better and positive results