Title: Predicting household wealth through neural network analysis of satellite imagery
Department: Data Science
Description: Today, climate change is having increasingly significant impacts on populations around the globe, making it essential to locate the populations that are the most vulnerable to the effects of climate change. However, it can be difficult to identify vulnerable populations, especially with censuses or large-scale surveys like the Demographic Health Survey (DHS). This means that many households and individuals are not accounted for and there are still significant periods of time between surveys. This project will explore labeled satellite images of DHS survey areas in South Asia. These images will be collected from Landsat and labeled using DHS data for India, Bangladesh, and Nepal. The goal of using these images is to predict the wealth index of the homes in each specific DHS cluster by training various neural network architectures. Ultimately, after training the models, this project will attempt to predict the wealth index of any area, given satellite imagery. If governments and aid organizations are able to detect areas with low wealth, they can identify the populations that require the most assistance to combat the effects of climate change.
Hometown: Baltimore, Maryland
Advisor: Daniel Runfola
Each William & Mary Honors Fellowships donation supports the whole Charles Center Honors Fellowships Fund. Donations inspired by specific research projects are distributed by the Charles Center in a way that benefits this project and others. To learn more about Honors Fellowships, please visit our website.