A dictionary is a reference data set that you can use to evaluate data in a mapping. Use dictionaries to verify that the data values on a data source or another object in a mapping are accurate and correctly formatted.
You can create a dictionary through the Data Quality or Reference 360 user interface or through the Dictionary API. You add a dictionary to an asset in Data Quality, and you add the asset to the corresponding transformation in a mapping in Data Integration.
For example, you can configure a rule specification to read a dictionary, and you can add the rule specification to a Rule Specification transformation in a mapping. You can also add a dictionary to a cleanse, parse, or labeler asset and add the asset to the corresponding transformation in a mapping.
When you run a mapping that contains the transformation, the mapping compares each input data value to the dictionary data and performs actions that you specify. For example, a rule specification might return a value that indicates a match between the input data value and the dictionary data. Or, a cleanse asset might return a preferred or correct version of the input value.
A dictionary can contain multiple columns. At least one column must contain the standard or preferred versions of a set of values that the organization uses. Other columns can contain alternative or related versions of the values, including versions that might appear in the source data. The column that contains the standard or preferred values is called the valid column. You can populate a dictionary with any combination of values that suits your organization and your project requirements.
Types of dictionary
You can create a dictionary in Data Quality or in Reference 360. You can also create a dictionary through the Dictionary API. All types of dictionary function in the same way when you add them to a data quality asset.
A dictionary that you create in Reference 360 or through the Dictionary API appears in the Explore page in the same manner as a dictionary that you create in Data Quality.
Consider that you are a data analyst in a retail organization. You find reference data in Reference 360 that can verify the gender values in a customer report. You create a dictionary from the data in Reference 360, and you add the dictionary to a labeler asset in Data Quality. You then add the asset to a labeler transformation in a Data Integration mapping and run the mapping to verify the gender data.
If you create a dictionary in Reference 360, you can view the contents of the dictionary in Data Quality but you cannot edit the contents in Data Quality. To edit the dictionary data, open the dictionary source in Reference 360.
For more information about how to create a dictionary in Reference 360, see Reference 360 in Reference 360 help.
For more information about how to create a dictionary through the Dictionary API, see Create a dictionary in the Data QualityAPI Reference.