Aggregating Secondary Source Data for Air Cargo Prescreening

We propose to improve the effectiveness of air cargo prescreening by enabling background knowledge about companies and products to be used for threat assessment. The Transportation Security Administration's (TSA) current approach for pre-screening air cargo shipments is based primarily on the Known Shipper Program, which has several shortcomings. By combining sophisticated data extraction and integration technology with state-of-the-art data mining capabilities, threat assessment rules can be developed to help identify high-risk cargo. However, threat assessment relies on having data about the entities being assessed. In this project, we propose to develop novel data aggregation methods to automatically gather information about companies and products from corporate web sites, business directories, and other internet sources. We can then augment primary data sources (cargo manifest, database of past cargo shipments, package characteristics such as weight and volume) with additional background data (shipper and receiver information, shippable goods information) to perform threat assessment, and thereby route high-risk cargo for additional inspection. The use of this background data has great potential to significantly improve the ability of the TSA to detect vulnerabilities that may arise in the shipment of air cargo to, from, and within the United States.