【摘要】 不同地区的用户兴趣不同,并且当推荐物品具有位置属性时,用户更加倾向于离自身较近的物品。根据用户和物品的位置信息来捕获用户兴趣能有效地提高个性化推荐精度。为了有效处理用户和物品的位置信息,在推荐系统中引入金字塔模型(PS)来实现用户分区和用户旅行代价的计算,提出了基于金字塔模型的协同过滤算法(PMCF),来生成对用户的Top-N物品推荐。使用MovieLens数据集、Foursquare数据集和Synthetic数据集来分别评估算法的有效性,实验表明,所提出的算法的准确度要高于传统的推荐算法。更多还原

【Abstract】 Users have different interests in different regions,and when recommended items are spatial,users tend to travel a limited distance when visiting these venues.Accurately capturing user preferences according to the users’ and items’ location can improve the precision in recommender systems.To effectively deal with users’ and items’ location information,this paper introduced Pyramid Model(PM)in recommender systems for realizing users’ partitioning and calculating travel penalty,and presented a collaborative filtering recommendation algorithm based on Pyramid model(PMCF)to generate Top-N recommend.MovieLens,Foursquare and Synthetic data set were quoted to evaluate the effectiveness of the algorithm.Experimental results show our algorithm has significant improvements in terms of effectiveness measured through precision.更多还原