Abstract:We present the hierarchical graph infomax (HGI) approach for learning urban region representations (vector embeddings) with points-of-interest (POIs) in a fully unsupervised manner, which can be used in various downstream tasks. Specifically, HGI comprises several key steps: (1) training category embeddings as the initial features of POIs; (2) interconnecting POIs with a graph structure and performing graph convolution to capture the uniqueness of each POI based on its spatial context; (3) aggregating POIs to the regional level using multi -head attention mechanisms, to consider the multi-faceted influence from POIs to regions; (4) performing graph convolution at the regional level to generate region representations, to incorporate the similarities between adjacent regions; (5) aggregating region representations to produce an embedding at the city level. The model is finally trained through maximizing the mutual information among the POI - region - city hierarchy, which facilitates the information from local (POIs) and global (city) scales flowing to the learned region representations, making them both locally and globally relevant. We perform extensive experiments on three downstream tasks, i. e., estimating urban functional distributions, population density, and housing price, in the study areas of Xiamen Island and Shenzhen, China. The results indicate that HGI considerably outperforms several competitive base-lines in all three tasks, which proves that HGI could produce meaningful and effective region representations. In addition, the learned region representations based on POIs can potentially be used for reinforcing data repre-sentations from other modalities, e.g., remote sensing data. The implementation of HGI can be found at htt ps://github.com/RightBank/HGI.

Beyond the First Law of Geography: Learning Representations of Satellite Imagery by Leveraging Point-of-Interests

Learning Representations of Satellite Imagery by Leveraging Point-of-Interests

A Spatial and Adversarial Representation Learning Approach for Land Use Classification with POIs

Mapping Human Perception of Urban Landscape from Street-View Images: A Deep-Learning Approach

Perceiving Commerial Activeness over Satellite Images

Mitigating Urban-Rural Disparities in Contrastive Representation Learning with Satellite Imagery

Towards Effective Next POI Prediction: Spatial and Semantic Augmentation with Remote Sensing Data

Predicting Multi-level Socioeconomic Indicators from Structural Urban Imagery

Geographic Mapping with Unsupervised Multi-Modal Representation Learning from VHR Images and POIs

POI Representation Learning by a Hybrid Model.

Beyond satellite imagery: the influence of map representations on socio-economic prediction

An Ensemble Learning Approach for Urban Land Use Mapping Based on Remote Sensing Imagery and Social Sensing Data

Self-Supervision, Remote Sensing and Abstraction: Representation Learning Across 3 Million Locations

Urban Perception of Commercial Activeness from Satellite Images and Streetscapes

Lightweight and Robust Representation of Economic Scales from Satellite Imagery

Learning Urban Region Representations with POIs and Hierarchical Graph Infomax

A Dynamic Network and Representation LearningApproach for Quantifying Economic Growth fromSatellite Imagery

Learning Geospatial Region Embedding with Heterogeneous Graph

Learning Spatiotemporal-Aware Representation for POI Recommendation

Multimodal Contrastive Learning of Urban Space Representations from POI Data

Simultaneous Extraction of Spatial and Attributional Building Information Across Large-Scale Urban Landscapes from High-Resolution Satellite Imagery