POI Recommendation with Federated Learning and Privacy Preserving in Cross Domain Recommendation

Point-of-Interest (POI) recommendation is one of the most popular recommendation methodologies. However, POI data is very sensitive and sparse. Users' reluctance to share their context information due to privacy concerns, along with the cold-start problem caused by data sparsity reduces recommendation efficiency. To address these issues, we propose a POI framework for cross-domain recommendation with federated learning and privacy protection features. It utilizes data in an auxiliary domain in users' interest analysis to alleviate the cold-start problem. Moreover, it applies federated learning by analyzing the users' historical data locally and encrypts latent feature distribution for knowledge migration to protect users' privacy. Experiments on real datasets have shown that our framework improves recommendation accuracy while preserving users' privacy as compared to convolutional neural network-based methods when analyzing users' comments.

I. Introduction

With the development of mobile localization technology, Point-of-Interest (POI) recommendations have attracted wide attention. POI recommendations can filter information based on users’ interests to achieve effective recommendation and greatly improve users’ experience while producing substantial economic benefits. Nowadays, many platforms support users releasing daily social network data while sharing their location "check-in" information and generating relevant comments. Unlike traditional product recommendations, people’s check-in data is more sensitive and significantly sparser in POI recommendations. Check-in data not only reflects a person's location, but also reveals users’ interests and other private details. For example, if a person frequently checks in to a hospital-related location over a period of time, we can infer that the user may have a health problem. In addition, POI recommendations have strict requirements on the time factor because people tend to checks in different POIs at different times during a day.