Learning Economic Indicators by Aggregating Multi-Level Geospatial Information


  • Sungwon Park KAIST
  • Sungwon Han KAIST
  • Donghyun Ahn KAIST
  • Jaeyeon Kim New York University
  • Jeasurk Yang National University of Singapore
  • Susang Lee KAIST
  • Seunghoon Hong KAIST
  • Jihee Kim KAIST
  • Sangyoon Park University of Hong Kong
  • Hyunjoo Yang Sogang University
  • Meeyoung Cha KAIST IBS




AI For Social Impact (AISI Track Papers Only), Computer Vision (CV)


High-resolution daytime satellite imagery has become a promising source to study economic activities. These images display detailed terrain over large areas and allow zooming into smaller neighborhoods. Existing methods, however, have utilized images only in a single-level geographical unit. This research presents a deep learning model to predict economic indicators via aggregating traits observed from multiple levels of geographical units. The model first measures hyperlocal economy over small communities via ordinal regression. The next step extracts district-level features by summarizing interconnection among hyperlocal economies. In the final step, the model estimates economic indicators of districts via aggregating the hyperlocal and district information. Our new multi-level learning model substantially outperforms strong baselines in predicting key indicators such as population, purchasing power, and energy consumption. The model is also robust against data shortage; the trained features from one country can generalize to other countries when evaluated with data gathered from Malaysia, the Philippines, Thailand, and Vietnam. We discuss the multi-level model's implications for measuring inequality, which is the essential first step in policy and social science research on inequality and poverty.




How to Cite

Park, S., Han, S., Ahn, D., Kim, J., Yang, J., Lee, S., Hong, S., Kim, J., Park, S., Yang, H., & Cha, M. (2022). Learning Economic Indicators by Aggregating Multi-Level Geospatial Information. Proceedings of the AAAI Conference on Artificial Intelligence, 36(11), 12053-12061. https://doi.org/10.1609/aaai.v36i11.21464