Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2511.01408 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908625224597504 |
|---|---|
| author | Pettersson, Markus B. Daoud, Adel |
| author_facet | Pettersson, Markus B. Daoud, Adel |
| contents | Accurate, fine-grained poverty maps remain scarce across much of the Global South. While Demographic and Health Surveys (DHS) provide high-quality socioeconomic data, their spatial coverage is limited and reported coordinates are randomly displaced for privacy, further reducing their quality. We propose a graph-based approach leveraging low-dimensional AlphaEarth satellite embeddings to predict cluster-level wealth indices across Sub-Saharan Africa. By modeling spatial relations between surveyed and unlabeled locations, and by introducing a probabilistic "fuzzy label" loss to account for coordinate displacement, we improve the generalization of wealth predictions beyond existing surveys. Our experiments on 37 DHS datasets (2017-2023) show that incorporating graph structure slightly improves accuracy compared to "image-only" baselines, demonstrating the potential of compact EO embeddings for large-scale socioeconomic mapping. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_01408 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Leveraging Compact Satellite Embeddings and Graph Neural Networks for Large-Scale Poverty Mapping Pettersson, Markus B. Daoud, Adel Machine Learning Accurate, fine-grained poverty maps remain scarce across much of the Global South. While Demographic and Health Surveys (DHS) provide high-quality socioeconomic data, their spatial coverage is limited and reported coordinates are randomly displaced for privacy, further reducing their quality. We propose a graph-based approach leveraging low-dimensional AlphaEarth satellite embeddings to predict cluster-level wealth indices across Sub-Saharan Africa. By modeling spatial relations between surveyed and unlabeled locations, and by introducing a probabilistic "fuzzy label" loss to account for coordinate displacement, we improve the generalization of wealth predictions beyond existing surveys. Our experiments on 37 DHS datasets (2017-2023) show that incorporating graph structure slightly improves accuracy compared to "image-only" baselines, demonstrating the potential of compact EO embeddings for large-scale socioeconomic mapping. |
| title | Leveraging Compact Satellite Embeddings and Graph Neural Networks for Large-Scale Poverty Mapping |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2511.01408 |