Tourism Interest Point Recommendation Algorithm Based on Hybrid Collaborative Filtering
A tourism point of interest(POI)recommendation algorithm H-S-LMF based on hybrid collaborative filtering,social fea-ture information,and logistic matrix decomposition is proposed to address the problems of insufficient exploration of users'potential interests and low recommendation accuracy in existing POI recommendation methods.The algorithm firstly obtains user preferences through a hybrid collaborative filtering algorithm,then measures the influence of social features on tourism interest point recommen-dations based on the similarity of common check-in behaviors among users and friendship factors between users,and finally inte-grates user preferences and social features into logistic matrix decomposition to improve the accuracy of recommendations.The ex-perimental results show that compared with the optimal baseline model,H-S-LMF achieves higher accuracy on Yelp and Gowalla da-tasets.The precision rate of H-S-LMF(Precision@10)increased by 69.53%and 63.23%respectively,and the recall rate(Recall@10)increased by 73.61%and 59.05%respectively.
万方数据
维普
