A data domain is a predefined or user-defined Model repository object that uses rules to discover the functional meaning of column data or column names. The data domain rules define data patterns and column name patterns that match source data and metadata. You can use the data domain rules to update the data domain logic.
Use the data domains in the Data Domains accelerator to discover the functional meaning of source data based on column names or column data values.
The Data Domains accelerator includes the following types of rule:
Data rule. Finds columns with data that matches the logic that the rule defines.
Column name rule. Finds columns with column names that match column-name logic that the rule defines.
The data domain rules return Boolean values that indicate whether the column data or column name meets the rule criteria. The data domain rules use regular expressions or reference tables to look for specific values or patterns. For example, you can use a nine-digit rule expression to find data values in the Social Security number format.
When you use expressions in data domain rules, some unrelated data values might also meet the rule expression criteria. For example, United States ZIP codes in the source data might meet the Social Security number format. To make the data domain inference effective, review the data domain discovery results for discrepancies. After you review and verify the data domain discovery results, you can decide to associate a data domain with a data column.