期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2016
卷号：5
期号：4
页码：1092-1095
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Due to the advent and popularity of the social network, peoples use to share their experiences like product reviews, ratings etc. This leads to use the new factors such as interpersonal influence and interpersonal interest similarity to solve the cold start and data sparsity problems in recommendation systems. Some factors are already used in existing systems but they have not considered fully. In this paper we are considering four social factors such as user personal interest, user interpersonal interest similarity, user social context and interpersonal influence, fused all these factors into unified recommendation model based on probabilistic matrix factorization. User personal interest factor is mainly used to recommend to the experienced users, and to solve cold start problem we considered interpersonal influence and user interpersonal interest, and we are considered social context to recommend the location specific products to the users.
关键词：Interpersonal influence; Social context; User ; interpersonal interest similarity; User personal interest
