Title :
Research on WEB Cache Prediction Recommend Mechanism Based on Usage Pattern
Author :
Lin Jianhui ; Huang Tianshu ; Yang Chao
Author_Institution :
Wuhan Univ., Wuhan
Abstract :
Cache prefetching technique can improve the hit ratio and expedite users visiting speed. After analyzed the recommend system in e-business, this paper studied the characteristics how user visit Web page and proposed a Web prefetching recommender system based on usage pattern. This system cluster user behavior through an improved ant colony algorithm, then usage pattern can be abstracted from these classes through sequence mining. These sequence patterns are applied to forecast the coming behavior of users thus improve the hit ratio of system. Experiment result proves the validity of the system.
Keywords :
Internet; data mining; information filtering; storage management; Web cache prediction; Web page; Web prefetching recommender system; ant colony algorithm; cache prefetching; e-business; hit ratio; sequence mining; usage pattern; Chaos; Clustering algorithms; Data mining; Information technology; Pattern analysis; Prefetching; Probability; Recommender systems; Uniform resource locators; Web pages;
Conference_Titel :
Knowledge Discovery and Data Mining, 2008. WKDD 2008. First International Workshop on
Conference_Location :
Adelaide, SA
Print_ISBN :
978-0-7695-3090-1
DOI :
10.1109/WKDD.2008.9
