|
The U.S. Customs and Border Protection (CBP) Arrival and Departure Information System (ADIS) consolidates data from a variety of systems to create a unique person-centric record with complete travel history. Originally, CBP created ADIS to identify individuals who had overstayed their class of admission (“visa overstays”); now CBP is broadening its use of ADIS for all traveler encounters to support a variety of non-law enforcement use cases.
The system matches data utilizing a series of custom and open source matching algorithms designed to facilitate exact and predictive matching capabilities. Damerau–Levenshtein distance is the open source algorithm integrated within ADIS for predictive “fuzzy” matching. Before the system matches information, business rules and data normalization standards are applied to augment the data into the system’s logical data model and to determine if the information meets requirements for data ingestion. Once the information is cleared for processing, a series of tier based matching steps are applied to identify possible match candidates or to create a new identity in the system. The majority of the system’s matches are made through exact matches.
|
