For applications that contain multiple data sets, refinement rules allow you to connect attributes from the different data sets.
When end users refine a data set by an attribute that belongs to a refinement rule, other data sets are refined by other attributes that belong to the refinement rule.
Refinement rules work best with attributes that have the same or similar set of values.
For example, a sales data set for automotive data includes the make and vehicle ID number (VIN) of cars that were sold. Another data set containing warranty claims also includes the make and VIN of cars for which warranty claims were filed.
If you then create refinement rules for the make and VIN attributes, then when users refine one data set by a make or VIN, the other data set also is refined by that make or VIN.
Date attributes are another type of attribute that benefit from refinement rules. For example, when refining a sales transaction data set by the sales date, the Twitter posts data set is refined by the post date.
When you add a data set to an existing application, you can have Studio automatically create refinement rules for attributes that have the same attribute name, data type, multi-value setting, and refinement behavior. You can then use the Refinement Rules page to edit and remove those rules.
A refinement rule can only contain one attribute from each data set.
