Abstract

Location-based social networks (LBSNs) have gained significant popularity nowadays and their location-sharing features promote social interactions and foster community formation. However, friend recommendation on LBSNs remains a challenging research problem. As check-in trajectories indicate user proximity, we propose a deep learning method to represent users and locations by mining user trajectories and generate top-k friend recommendation.

Share

COinS
 

Distributed Representations of Users and Locations for Friendship Recommendation on Location-Based Social Network

Location-based social networks (LBSNs) have gained significant popularity nowadays and their location-sharing features promote social interactions and foster community formation. However, friend recommendation on LBSNs remains a challenging research problem. As check-in trajectories indicate user proximity, we propose a deep learning method to represent users and locations by mining user trajectories and generate top-k friend recommendation.