Agency Interoperation for Effective Data Mining in Border Control and Homeland Security Applications

US Customs embarked on a major modernization initiative of its Information Technology systems. Drawing in data from Customs trade systems, targeting inspectors review manifest information as well as strategic and tactical intelligence to determine which shipments and containers are “high-risk.” This entails a considerable level of communication and data sharing between various government agencies. Our NSF funded project, funded through the digital government program, aims at providing decision makers with the ability to extract and fuse information from multiple, heterogeneous sources in response to a query while operating under a decentralized security administration. Based on the idea of “Smart Borders”, the system will utilize data available from different agencies, ports and customs divisions to supplement the profiling by targeting towards anomalies, and detect various flags raised by non-conforming shipments or abnormal behavior of inbound cargos and raise a combination of alerts. This project proposes generalizable development work, which devises solutions to accomplish secure interoperation among different government agencies. The output of this project would ideally enhance the security aspect of the Automated Commercial Environment (ACE) system by incorporating the concept of semantic interoperability, anomaly detection and subsequent spatial and geographical visualization of information that can help Customs inspectors make better decisions. This project is a collaboration between the industry, government and the academia, providing the opportunity to directly influence the practical needs of the government agency, in this case the US Customs.