Behavior-Based Location Recommendation on Location-Based Social Networks

Abstract

Location recommendation makes suggestions of nearby locations based on user’s locational preferences and spatial movement patterns. In this paper, we propose two novel location recommendation methods called Behavior Factorization (BF) and Latent Behavior Analysis (LBA). Both methods utilize behavioral and spatio-temporal patterns in user movements to make location recommendation. Experiments on a real-world dataset shows that the proposed methods outperform existing location recommendation methods in terms of both precision and recall. Comparing LBA and BF, it is observed that LBA achieves better results since it utilizes the number of times each pattern has happened in the dataset.