How can we create a recommender system for tourism? A location centric spatial binning-based methodology using social networks
Malika Acharya, Shilpi Yadav, Krishna Kumar Mohbey
Abstract
Point of Interest (POI) recommendation is an important Location-Based Social Network (LBSN) task that has become a hotspot in the past decade. It aims to exploit the user's preferences for venues and recommend the POIs for their next visit. In the past, several works using geographical, temporal, social, and other contextual information have been brought to the forefront. But the data sparsity problem has foiled these attempts, and thereby they could not furnish the required level of accuracy. Here in this paper, we propose Long Short-Term Memory (LSTM) based method for POI recommendation. The crux of the approach is the spatial binning used to group the venues in proximity to the user's previously visited venues. This approach has been evaluated on three real-world databases, i.e., Gowalla, Foursquare NYC, and Foursquare TKY. Its evaluation suggests its efficacy. Further, its high accuracy and less time consumption supersede all the other approaches.