Dynamic discovery of favorite locations in spatio-temporal social networks
作者：
Highlights：
• An original framework of semi-supervised learning is presented for POI recommendation.
• Multiple factors are jointly studied to excavate complex spatio-temporal patterns of visiting behaviors.
• A network embedding method is proposed to learn the vectors of users and POIs in an embedding space with low dimensionality.
• A dynamic factor graph model is proposed to model different factors including the correlation of users’ vectors and POIs’ vectors.
摘要
•An original framework of semi-supervised learning is presented for POI recommendation.•Multiple factors are jointly studied to excavate complex spatio-temporal patterns of visiting behaviors.•A network embedding method is proposed to learn the vectors of users and POIs in an embedding space with low dimensionality.•A dynamic factor graph model is proposed to model different factors including the correlation of users’ vectors and POIs’ vectors.
论文关键词：Location-based social networks,POI recommendation,Heterogeneous networks,Factor graph model,Network embedding
论文评审过程：Received 8 March 2020, Revised 19 May 2020, Accepted 6 June 2020, Available online 27 June 2020, Version of Record 27 June 2020.
论文官网地址：https://doi.org/10.1016/j.ipm.2020.102337
