Open AccessPoint of Interest Recommendation Based on Graph Convolutional Neural Network
Author(s) -
Fubo Zhai,
Baozhu Li
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1883/1/012132
Subject(s) - computer science , point of interest , popularity , convolutional neural network , graph , recommender system , the internet , artificial intelligence , information retrieval , data mining , machine learning , world wide web , data science , theoretical computer science , psychology , social psychology
The rapid development of the mobile Internet makes location-based social networks (LBSNs) play an increasingly important role in practical applications. Among them, point of interest(POI) recommendation is a research hotspot in the current context. As a kind of graph data, social network can naturally express the data structure in real life. In view of the current POIs recommendation research ignoring the diversity of graph data, we proposed a POI recommendation based graph convolutional neural network (PBGCN) model, which used the check-in information, popularity characteristics of interest points, and users’ social behaviors to recommend interest points through graph convolutional neural networks(GCN). Compared with other latest recommendation methods, our model has improved accuracy. This proves the feasibility of GCN in point of interest recommendation.