HGAurban: Heterogeneous Graph Autoencoding for Urban Spatial-Temporal Learning

TL;DR

HGAurban introduces a self-supervised heterogeneous graph autoencoder that effectively captures complex spatial-temporal dependencies in noisy urban data, improving urban sensing applications like traffic and crime prediction.

Contribution

It presents a novel masked autoencoder framework leveraging heterogeneous graph encoding and self-supervised learning for robust urban spatial-temporal data representation.

Findings

Outperforms state-of-the-art methods in multiple tasks

Robustly handles noise and sparsity in real-world data

Enhances modeling of dynamic temporal correlations

Abstract

Spatial-temporal graph representations play a crucial role in urban sensing applications, including traffic analysis, human mobility behavior modeling, and citywide crime prediction. However, a key challenge lies in the noisy and sparse nature of spatial-temporal data, which limits existing neural networks' ability to learn meaningful region representations in the spatial-temporal graph. To overcome these limitations, we propose HGAurban, a novel heterogeneous spatial-temporal graph masked autoencoder that leverages generative self-supervised learning for robust urban data representation. Our framework introduces a spatial-temporal heterogeneous graph encoder that extracts region-wise dependencies from multi-source data, enabling comprehensive modeling of diverse spatial relationships. Within our self-supervised learning paradigm, we implement a masked autoencoder that jointly processes node features and graph structure. This approach automatically learns heterogeneous spatial-temporal patterns across regions, significantly improving the representation of dynamic temporal correlations. Comprehensive experiments across multiple spatiotemporal mining tasks demonstrate that our framework outperforms state-of-the-art methods and robustly handles real-world urban data challenges, including noise and sparsity in both spatial and temporal dimensions.