期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2016
卷号：4
期号：1
页码：153
DOI：10.15680/IJIRCCE.2016.0401029
出版社：S&S Publications
摘要：The problem of point of interest recommendation is used to provide personalized recommendations of places like malls, hotels, movie theatres etc. Now a days, the increasing prevalence of mobile devices and of location based social networks (LBSNs) provides significant new opportunities as well as challenges. To choose POI is complex for user to make decision because of factors, such as own preferences, geographica l considerations, and user mobility behaviors . It is difficult because of connecting LBSNs to the mobile devices. In previous studies regarding to POI recommendations, it has been seen that, there is lack an integrated analysis of the joint effect of multi ple factors. But here, the model is effective and widely used for POI recommendations which require consideration of the unique characteristics of LBSNs. At the end, we propose a general geographical probabilistic factor model framework which takes variou s factors into consideration. Propose framework allows capturing the geographical influences on a user's check - in behavior and user mobility behaviors can be effectively leveraged in the recommendation model. After that based on our Geo - PFM framework, we f urther develop a Poisson Geo - PFM which provides a probabilistic generative process for the entire model and is effective in user check - in data as implicit feedback for better POI recommendations. Finally, experimental results on three real - world LBSN datas ets show that the proposed recommendation methods outperform state - of - the - art latent factor models by a significant margin
关键词：Recommender systems; point of interest (POI); probabilistic factor model; location ; - ; based social ; networks
