For base objects with the fuzzy match/search strategy, the match process uses standard population sets to account for national, regional, and language differences. The population set affects how the match process handles tokenization, the match / search strategy, and match purposes.

A population set encapsulates intelligence about name, address, and other identification information that is typical for a given population. For example, different countries use different address formats, such as the placement of street numbers and street names, location of postal codes, and so on. Similarly, different regions have different distributions for surnames—the surname “Smith” is quite common in the United States population, for example, but not so common for other parts of the world.

Population sets improve match accuracy by accommodating for the variations and errors that are likely to appear in data for a particular population.