Empowering Next POI Recommendation with Multi-Relational Modeling

TL;DR

This paper introduces MEMO, a novel framework that leverages multi-relational data and inter-temporal user-POI influences to significantly improve next POI recommendation in location-based social networks.

Contribution

The paper proposes MEMO, a multi-network representation learning framework that explicitly models heterogeneous relations and temporal influences for better POI prediction.

Findings

MEMO outperforms state-of-the-art methods on real-world datasets.

Heterogeneous relational modeling improves recommendation accuracy.

Temporal user-POI influence modeling enhances personalization.

Abstract

With the wide adoption of mobile devices and web applications, location-based social networks (LBSNs) offer large-scale individual-level location-related activities and experiences. Next point-of-interest (POI) recommendation is one of the most important tasks in LBSNs, aiming to make personalized recommendations of next suitable locations to users by discovering preferences from users' historical activities. Noticeably, LBSNs have offered unparalleled access to abundant heterogeneous relational information about users and POIs (including user-user social relations, such as families or colleagues; and user-POI visiting relations). Such relational information holds great potential to facilitate the next POI recommendation. However, most existing methods either focus on merely the user-POI visits, or handle different relations based on over-simplified assumptions while neglecting relational heterogeneities. To fill these critical voids, we propose a novel framework, MEMO, which effectively utilizes the heterogeneous relations with a multi-network representation learning module, and explicitly incorporates the inter-temporal user-POI mutual influence with the coupled recurrent neural networks. Extensive experiments on real-world LBSN data validate the superiority of our framework over the state-of-the-art next POI recommendation methods.