Predicting cross-border migration patterns
Modeling push-and-pull factors in cross-border migration with deep learning
Human migratory decisions are the result of a complex range of interacting factors, including economic, social and environmental vulnerabilities. Forming a clear understanding why, how, and where migration occurs across U.S. borders is critical for the Department of Homeland Security (DHS) to increase the security of these expansive areas.
This project will develop a computational framework, analytical capabilities, and modeling of cross-border migration to enhance national security. Through a combination of data and innovative software tools, DHS officials will gain new insights for the efficient allocation and deployment of agency resources to increase border security.
Through extensive research, the team will work to fully understand the complex and interconnected challenges of cross-border migration, including identifying and analyzing potential sources of the problem. This includes looking at downward trends of economic conditions of various regions and monitoring of migration activity to help predict future border migration events. The team will observe the gentrification and/or deterioration of certain neighborhoods in a particular city can also affect migration patterns.
This project will study and compile extensive data to develop a complex computational framework, innovative analytical capabilities, and software to assist DHS in the agency’s mission. These tools will support DHS agencies such as U.S. Customs and Border Protection (CBP) to aggregate and better understand migration data. With these new assets, agency officials will be empowered to better manage border operations through greater insights into resource allocation and adjustments to response measures.
The framework and analytical capabilities developed through this project may eventually be used in a variety of broader DHS operations, for activities monitoring drug trafficking routes within and across countries or observing gang operations in Central American and U.S. communities.
Research Leadership Team
Principal Investigator: Anthony Stefanidis, College of William and Mary
Co-PI: Daniel Miller Runfola, College of William and Mary