SAS Quality Knowledge Base for Contact Information 26

Identification Analysis Definitions

Identification analysis definitions specify data and logic that can be used to identify the semantic type of a data string. For example, an identification analysis definition might be used to determine whether a certain string represents the name of an individual or an organization.

Identification analysis definitions can be useful when you want to perform analytics on a field in a database table. For instance, you might want to determine what percentage of fields contain addresses, and what percentage contains organization names. Identification analysis definitions are also useful if you want to programmatically analyze a data string before applying a different type of definition. For example, you might use an identification analysis definition to determine whether a string represents an address or the name of a company. Then you can apply the Address match definition to all strings that represent addresses.

The output of an identification analysis definition is a string that describes the semantic type of the data. There could be two or more possible outputs for an identification analysis definition.

Following is an example of results that you might see when using an identification analysis definition.