Collaborative Research: HNDS-I. Mobility Data for Communities (MD4C): Uncovering Segregation, Climate Resilience, and Economic Development from Cell-Phone Records

Cell phone location records, called mobility data (MD), are data about where people go. These data are an essential resource to study how people move and interact in society. However, these data are also very complex, so very few scientists have the technical skills to work with them. There are also privacy or data ownership issues that require scientists to have permission to access them. For these reasons, policymakers, practitioners, and the people in the communities that the data describe do not use them despite their importance. This award supports the creation of Mobility Data for Communities (MD4C), an infrastructure project that converts MD into a series of useful measures that describe mobility across a region's neighborhoods and places within those neighborhoods. These measures are accessible to everyone, including researchers, policymakers, and practitioners. MD4C’s tools are designed to resolve biases in the measures that result from not having certain people (for example, the elderly or unhoused who use cell phones less frequently) or having too many other people (for example, younger professionals who use cell phones all day) in the datasets. Bias is measured and corrected by comparing MD to data from a survey of people who voluntarily share their mobility information. Once corrected, deep learning techniques can extract measurements of mobility patterns related to social equity: racial and socioeconomic segregation, economic development, and climate resilience. The MD4C platform makes these bias-adjusted mobility measurements accessible for researchers and others and makes new research possible across the social sciences. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.