Multi-Graph Fusion Networks for Urban Region Embedding

TL;DR

This paper introduces multi-graph fusion networks (MGFN) that integrate various mobility graphs and employ cross-attention to learn comprehensive urban region embeddings, significantly improving prediction tasks like crime detection.

Contribution

The paper proposes a novel multi-graph fusion framework with a cross-attention mechanism for enhanced urban region embedding from mobility data.

Findings

MGFN outperforms state-of-the-art methods by up to 12.35%

Effective integration of spatio-temporal graphs improves embedding quality

Cross-attention mechanism captures complex mobility patterns

Abstract

Learning the embeddings for urban regions from human mobility data can reveal the functionality of regions, and then enables the correlated but distinct tasks such as crime prediction. Human mobility data contains rich but abundant information, which yields to the comprehensive region embeddings for cross domain tasks. In this paper, we propose multi-graph fusion networks (MGFN) to enable the cross domain prediction tasks. First, we integrate the graphs with spatio-temporal similarity as mobility patterns through a mobility graph fusion module. Then, in the mobility pattern joint learning module, we design the multi-level cross-attention mechanism to learn the comprehensive embeddings from multiple mobility patterns based on intra-pattern and inter-pattern messages. Finally, we conduct extensive experiments on real-world urban datasets. Experimental results demonstrate that the proposed MGFN outperforms the state-of-the-art methods by up to 12.35% improvement.