为了提高兴趣点(point of interest,POI)推荐的准确性和个性化,提升用户对推荐结果的满意度,针对不同活跃度用户的特点,提出一种融合用户活跃度的上下文感知兴趣点推荐算法(A POI recommendation algorithm that integrates geographical,categorical,and temporal factors,while simultaneously considering user activity),简称AU-GCTRS.首先,为缓解数据稀疏性和冷启动问题,引入多维上下文信息;其次,通过挖掘用户签到频率、签到兴趣点数量和签到时间,将用户划分为不同活跃度的群体;最后,综合用户活跃度与上下文分数,将得分高的前K个兴趣点推荐给用户.在真实数据集上进行实验表明,AU-GCTRS算法比其他流行算法更有效地缓解了数据稀疏性和冷启动问题,提高了推荐准确率和召回率.
An algorithm integrating user activity into a context-aware point-of-interest recommendation
To enhance the accuracy and personalization of POI(point of interest)recommendations,as well as to improve user satisfaction with recommended results,a context-aware POI recommendation algorithm that integrates geographical,categorical,and temporal factors while considering user activity,called AU-GCTRS,is proposed in this study.It firstly integrates multidimensional context information into the recommendation system to alleviate data sparsity and cold-start issues,then provides personalized recommendations for users with different levels of activity by mining the frequency of user check-in,quantity of POIs,and check-in time.Finally,by integrating user activity and context scores,the Top-K POIs are recommended to users.Experiments have been conducted on the New York City real check-in dataset collected in Foursquare.The results show that compared with other popular POI recommendation algorithms,the AU-GCTRS algorithm can effectively alleviate data sparsity and cold-start problems,and has a certain improvement in recommendation accuracy and recall.
point of interest recommendationuser activitycontext awaregeographical scoretime scorecategory-popularity scorecollaborative filteringcheck-in data
万方数据
维普
