The U.S. Customs and Border Protection (CBP) employs AI-enhanced matching algorithms for entity resolution

Summary:

The U.S. Customs and Border Protection (CBP) department utilizes an AI tool to enhance risk assessment processes by analyzing trade and travel data in real-time.

Client: 

U.S. Customs and Border Protection (CBP), part of the  United States Department of Homeland Security (DHS)

Problem Statement: 

The Department of Homeland Security (DHS) oversees border security through the Customs and Border Protection (CBP), the largest law enforcement agency in the U.S. The CBP monitors activities along the country’s land, air, and sea borders, identifying suspicious individuals. With 328 ports of entry and nearly a million people crossing them daily, this is a significant and demanding task. The CBP aimed to develop and launch the ‘Global Travel Assessment System (GTAS)’ in response to UN Security Council Resolution 2178 on Foreign Terrorist Fighters. Consequently, the adoption of ML and AI technology by DHS was inevitable.

Results: 

• The system generates match probabilities ranging from 0 to 1, aiding human managers in making informed decisions. 
• The software enables GTAS to make matches in under five seconds, significantly enhancing the speed at which users can generate predictive risk models for incoming travelers.

AI Solution Overview:

The Department of Homeland Security employs AI-enhanced entity resolution in its Global Travel Assessment System (GTAS). Designed by U.S. Customs and Border Protection (CBP) as an open-source platform, GTAS receives and stores standard air traveler information (Advanced Passenger Information (API) and Passenger Name Record (PNR)) to enable real-time risk modeling. 

Tamr provided CBP with machine learning-enhanced software to analyze API/PNR data within GTAS, improving entity identification and matching at security and border checkpoints. This increases the speed at which GTAS can confirm the identity of “trusted travelers” or identify persons of interest. 

The free-to-use, web-based application, available on GitHub, efficiently screens travelers using standard API and PNR data. AI methods applied to CBP data support informed decision-making, with tools continuously evaluated for accuracy and precision, reinforcing CBP’s core mission within a layered risk assessment strategy.

References: 

1. Department of Homeland Security Artificial Intelligence Use Case Inventory. https://www.imwong.com/2023/12/13/department-of-homeland-security-artificial-intelligence-use-case-inventory/
2.  The Department of Homeland Security Uses AI-Enhanced Entity Resolution for its Global Travel Assessment System (GTAS). https://emerj.com/ai-case-studies/the-department-of-homeland-security-uses-ai-enhanced-entity-resolution-for-its-global-travel-assessment-system-gtas/
3. AI at the US Department of Homeland Security – Current Projects. https://emerj.com/ai-sector-overviews/artificial-intelligence-homeland-security/

Industry: Public Services (Military and Defense)

Vendor:  TAMR

Client: U.S. Customs and Border Protection (CBP), part of the United States Department of Homeland Security (DHS)

Publication Date: 2018

Key Words: AI, AI-enhanced matching algorithms, entity resolution, border protection