Enhancing the performance of web proxy server through cluster based prefetching techniques

Web caching is used to reduce the network traffic by caching web pages at the proxy server level. Nowadays caching alone is not sufficient because of World Wide Web has evolved rapidly from a simple information-sharing mechanism. This mechanism offer only static text and images to a rich assortment of dynamic and interactive services, such as video/audio conferencing, e-commerce and distance learning. Web is demanding to improve the cache performance. If we use the prefetching technique with caching then the performance of cache is improved. Prefetching fetches objects that are likely to be accessed in the near future and store them in advance thus the response time of the user request is reduced. In this paper, our main objective is to give a new framework to improve performance of web proxy server using web usage mining and prefetching scheme. Further, we cluster the user according to their access pattern and usage behavior with the help of K-Means algorithm and then Apriori algorithm is applied to generate rules for prefetching pages. This cluster based approach is applied on proxy server web log data to test the results using LRU and LFU prefetching schemes.