A match candidate group is a set of records in your input data that have something in common. For example, a match candidate group might consist of records containing the same postal code, city, or window key.
When you create a match candidate group you filter the records in your data, creating a subset of records that will be compared with each other in the window match. The original data is still available and is displayed on the Input Data tab; the candidate group will be used to perform the match and is displayed on the Match Candidates tab.
To use a match candidate group
-
In the Input Data pane, select Use Match Candidates.
-
Click the Define Filter icon . The Window Matching Row Filter window opens.
-
Select whether this will be an ‘and’ or an ‘or’ comparison.
-
Select data values to be used to filter the data. For example, to limit the data in the matching window to those records containing the names "John" or "Jane" or that include the last name "Nicholson" enter the following values:
|
Value Count
|
FIRSTNAME
|
LASTNAME
|
|
Value 1
|
John
|
|
| Value 2 |
Jane |
|
| Value 3 |
|
Nicholson |
If you require a more precise filter, you can add rows by clicking New Row.
-
Click OK. Records meeting the filter criteria will be displayed on the Match Candidates tab. When you perform a match, only these candidate records will be compared with each other to find matches.