CBP, Johns Hopkins Working on New Import Data Targeting Project
CBP's Commercial Targeting Enforcement Directorate and the Johns Hopkins University Applied Physics Laboratory (JHU/APL) are working together on a project to use "commercial trade data to enhance and identify pattern identification, entity links, and anomalies within large datasets," the Department of Homeland Security said. The project was mentioned within a newly released DHS 2016 data mining report to Congress (here). The project, called Socrates, "was initiated to determine the analytical abilities JHU/APL could apply to trade data analytics" and "during Fiscal Year (FY) 2016, a pilot test was completed by JHU/APL using import data," DHS said.
Sign up for a free preview to unlock the rest of this article
Export Compliance Daily combines U.S. export control news, foreign border import regulation and policy developments into a single daily information service that reliably informs its trade professional readers about important current issues affecting their operations.
The next step is to validate the pilot's results during FY 2017, DHS said. It said the validation consists of "CBP analysts or subject matter experts reviewing and determining whether the analytic results compare to past CBP findings, or provide additional recommendations for further review." Also, "decisions about individuals or entities, normally referred to as Importers of Record, may be made during the validation phase in order to test the results of analytics as they are developed; any decisions would involve requests for examination or document review."
The project uses a JHU/APL mathematician-designed program, named Socrates, "to develop algorithms to analyze large datasets looking for both normal and abnormal trade patterns of behavior," DHS said. "This results in the identification of anomalies in trade patterns or behaviors that may indicate illicit or criminal behavior in the trade environment. An anomaly may also indicate behavior that is completely within the law and is legal; though, it may not fit within the normal trade behavior of the dataset. These anomalies lead to the examination of real time importations, matching against the anomalies, to determine if a violation of law or illicit activity occurred. Initial test results performed on import data provided positive results of trade anomalies that will be validated for illicit trade activity during the next test phase in FY2017."
The data used for the pilot came from CBP's historical data, which "is extracted" from ATS and ACE monthly, DHS said in the report. The Socrates program "does not interact directly with CBP’s systems, but is housed on CBP servers connected to CBP’s network" and only CBP-cleared JHU/APL team members can access the data, it said. The CBP Office of International Trade "will be implementing additional security measures, such as the capability to track more detailed user activity, in future enhancements," DHS said. The annual DHS report on data mining discusses other targeting efforts that were included in past reports, such as the Automated Targeting System (ATS), but this is the first time Socrates was included, DHS said.