MobilityNet: Open-Source Cyberinfrastructure for Cross-Domain Mobility Data

Human mobility data obtained from urban infrastructure (such as cellular, transportation, and payment networks) have the potential for significant scientific and societal impacts via research in multiple Science and Engineering (S&E) disciplines such as Computer Science, Transportation Engineering, Urban and Regional Planning, Geography, Epidemiology, and Economics. Their potential use cases include designing analytical, predictive, and intervention models for transportation/urban planning, autonomous driving, 5G deployment, emergency response, pandemic mitigation, etc. However, most of these mobility data are often proprietary and thus cannot be accessed or costly to acquire by the S&E community unless released by data owners.

The goal of this project is to design an innovative behavior-inspired generative CI tool MobilityNet for trustworthy mobility data synthesis (i.e., balancing utility, privacy, and fairness) to make the resultant synthetic data application-ready via the following three CI components built upon validated models (such as generative ML and differential privacy) from both technical and human aspects: (1) cross-domain real data curation to address data bias, (2) socially-informed real data interpretation to address data implicitness, and (3) privacy-preserving synthetic data generation to address data sensitivity. The resultant synthetic data will be used to support three categories of sequential yet fundamental S&E use cases (analytics, prediction, and intervention) for different disciplines.

See details.