A page prefetching technique utilizing semantic information of links

摘要

A large growth of web uses on the Internet causes latency perceived by the user. Prefetching is a good solution to this problem. Web page content provides the meaningful data to predict the coming requests in future. This paper presents a semantic prefetching technique which uses the content of the web page. The proposed technique works on the semantic preferences of the anchor text associated with the URL. This technique also uses the semantic information, explicitly embedded with each link, for more accurate predictions. It also computes the semantic association in order to improve the prediction accuracy. This prefetching technique would be more effective for long sessions and will achieve good hit rate.