Remote Sensing Predicts Long-term Indicators of Governance, Stability, and Well-being

• Main Authors: Irvine, John M., Angelini, Brigid, Crystal, Michael
• Format: Conference Proceeding
• Language: English
• Subjects: Biological system modeling; economic analysis; Economic indicators; Feature extraction; governance; Government; imagery; Machine learning; remote sensing; Satellites; social indicators; Soft sensors
• ISSN: 2332-5615
• DOI: 10.1109/AIPR52630.2021.9762189

Understanding a region's socio-economic conditions can inform the development of policies in both the public and private sectors. Commercial satellite imagery provides a potential avenue for abstracting socio-economic context in a quick and relatively inexpensive manner. Satellite images contain infrastructural and agricultural information that, in a previous study focused on Afghanistan and Botswana, provided a useful for characterizing regional socio-economic information. Previous studies have compared survey responses to imagery feature, using supervised machine learning models. Building on previous work, this study explores long-term assessments of a country. As with previous studies, image features extracted from commercial imagery form the explanatory variables for our models. In this case, however, we seek to predict annual indicators of conditions for a country as assessed by the World Bank. The models relate imagery-derived features to indicators of Political Stability, Control of Corruption, Rule of Law, Government Effectiveness, Voice and Accountability, and Gross Domestic Product using data from multiple countries.