Location embeddings improve geographic data analysis and prediction tasks
LEON: Location Embeddings from OSM Neighborhoods via Hexagonal Graph Masked Autoencoders
Machine Learning
Summary
Understanding the layout and features of places on Earth is important for many tasks like predicting housing prices or crime. The researchers created LEON, a new technique that learns about geographic areas by looking at patterns of amenities using maps divided into hexagonal grids. This method helps computers better recognize and represent different places from map data without needing direct labels. Tests show LEON improves performance in tasks such as identifying satellite images and predicting real-world geographic outcomes.
What this means in practice
- •For urban planners: Use geographic embeddings to predict urban indicators like housing prices and crime from map-based amenity patterns.
- •For satellite imagery analysts: Improve classification of satellite images by integrating spatial embeddings derived from map features.
Authors
Szymon Soltysiak, Radoslaw Malek, Jedrzej Kusnierz, Milosz Chojecki, Piotr Szymanski, Aleksandra Kawala-Sterniuk
Abstract
Geographic information systems increasingly rely on sophisticated spatial representation learning techniques to extract meaningful patterns from complex geospatial data. This paper introduces LEON, a novel self-supervised framework that adapts Graph Masked Autoencoders (GraphMAE) for geospatial region representation learning. Our method leverages the inherent spatial structure of geographic data by constructing hexagonal grid graphs using H3 indexing and applying masked autoencoding techniques to learn robust spatial embeddings from OpenStreetMap (OSM) amenity distribution patterns. We evaluate LEON on multiple real-world datasets including EuroSAT satellite imagery classification and various geographic prediction tasks (housing prices, crime prediction, and urban analytics). Experimental results demonstrate that LEON achieves significant improvements in spatial understanding, with up to 1.87% accuracy improvement on EuroSAT classification and consistent performance gains across geographic prediction benchmarks. The learned embeddings exhibit highly structured and distinct properties, making them particularly suitable for downstream spatial analysis tasks. Our findings suggest that self-supervised learning provides an effective paradigm for geospatial region representation learning using widely available crowdsourced data.