期刊名称：International Journal of Advanced Research In Computer Science and Software Engineering
印刷版ISSN：2277-6451
电子版ISSN：2277-128X
出版年度：2013
卷号：3
期号：3
出版社：S.S. Mishra
摘要：Recommendation systems help the people to find things according to their interest and they are used widely with the development of electronic commerce. Many recommendation systems employ the collaborative filtering (CF) techniques, which are proved to be one of the most successful in recommender systems in recent years. CF techniques can be applied in entertainment domain also, for example in Amazon.com. CF provides recommendation based on the individual preferences. Because of the obvious increase of products in electronic commerce systems, the time consumed in searching the real time requirement of the customer sometimes results in failure. At the same time it suffers from poor novelty in case the user database increases. Less novelty is the major reason of increase in user frustration at times when he is looking for something new. To solve this problem in collaborative based filtering techniques, this paper propose a technique that enhance the novelty in a data set using the item cluster matrix and user cluster matrix, which contains item to user relationships. This technique is more efficient and accurate than the traditional one.
关键词：Recommender Systems; Collaborative based filtering; Novelty; User cluster; Item cluster
