Doing More with Less: Overcoming Data Scarcity for POI Recommendation via Cross-Region Transfer

TL;DR

This paper introduces Axolotl, a transfer learning method that leverages data-rich regions to improve POI recommendation accuracy in data-scarce regions through meta-learning and cluster-based transfer, with extensive cross-region experiments.

Contribution

The paper presents Axolotl, a novel transfer learning framework combining meta-learning and unsupervised clustering to enhance POI recommendations in regions with limited data, without requiring overlapping users.

Findings

Achieves up to 18% better recommendation performance than existing methods.

Effective transfer of location preferences across regions without user overlap.

Demonstrates robustness across diverse datasets from multiple countries.

Abstract

Variability in social app usage across regions results in a high skew of the quantity and the quality of check-in data collected, which in turn is a challenge for effective location recommender systems. In this paper, we present Axolotl (Automated cross Location-network Transfer Learning), a novel method aimed at transferring location preference models learned in a data-rich region to significantly boost the quality of recommendations in a data-scarce region. Axolotl predominantly deploys two channels for information transfer, (1) a meta-learning based procedure learned using location recommendation as well as social predictions, and (2) a lightweight unsupervised cluster-based transfer across users and locations with similar preferences. Both of these work together synergistically to achieve improved accuracy of recommendations in data-scarce regions without any prerequisite of overlapping users and with minimal fine-tuning. We build Axolotl on top of a twin graph-attention neural network model used for capturing the user- and location-conditioned influences in a user-mobility graph for each region. We conduct extensive experiments on 12 user mobility datasets across the U.S., Japan, and Germany, using 3 as source regions and 9 of them (that have much sparsely recorded mobility data) as target regions. Empirically, we show that Axolotl achieves up to 18% better recommendation performance than the existing state-of-the-art methods across all metrics.