AO Match Reduce Candidate Set
Overview
AO Match Reduce Candidate Set is used when you want to match a small number of records against a large number of records. For example:
Matching an update list against a master database.
Matching a customer list against a demographic universe.
Note that AO Match Reduce Candidate Set is designed for cases where the large record set is stored in a flat-file source. If your large record set is stored in a relational database or indexed DLD file, use the RDBMS Input tool's Dynamic Query feature or the Query Indexed Table instead.
AO Match Reduce Candidate Set compares a key on both inputs, and extracts those records from the large input that match the keys of one or more records from the small input, improving the performance of downstream matching processes. The key should be identical to the segmentation key used in the matching process itself. (See the documentation on the various matching macros for details on how to compute segmentation keys).
If the selected candidate flag field already exists and contains the value "Y" it will also have the value "Y" on output. This lets you chain together multiple Reduce Candidate Set macros and have the result be the logical "OR" of all the conditions. For example, if you have multiple disjoint match criteria like Name/Address, Email, and Name/SSN, you could chain together multiple AO Match Reduce Candidate Set macros to flag the candidates matching any of these criteria.
This macro has two connectors labeled "L" and "S" for the Large and Small record sets, respectively.
To see an example of this macro in action, open the AO Matching/Samples folder and then open the Reduce Candidate Set Sample project.
AO Match Reduce Candidate Set configuration parameters
AO Match Reduce Candidate Set has a single set of configuration parameters:
Input mapping
Parameter | Definition |
|---|---|
Small file key | Required. Key field from Small input. Default: Blank. |
Large file key | Required. Key field from Large input. Default: Blank. |
Output mapping
Parameter | Definition |
|---|---|
Candidate flag | Required. Appended to records from the Large input, this flag contains "Y" for selected candidates and "N" for rejected candidates. If this field already exists and contains the value "Y" it will also have the value "Y" on output. Default: CANDIDATE_FLAG |
Options
Parameter | Definition |
|---|---|
Remove non-candidates | Select this to remove the non-candidates (those with the Candidate Flag set to "N"). If you have multiple candidate-selection criteria, you can combine the results of multiple AO Match Reduce Candidate Set macros (say, one for name/address, one for email, and one for phone). In that case, leave this cleared and select the candidates in a post-filtering step. Default: No. |
Configure AO Match Reduce Candidate Set
To configure AO Match Reduce Candidate Set
On the canvas, connect the smaller input to the AO Match Reduce Candidate Set macro's "S" connector, and connect the larger input to the "L" connector.
Select the AO Match Reduce Candidate Set icon, and then go to the Configuration tab on the Properties pane.
Select Small file key and select the field containing the segmentation key for the "S" input.
Select Large file key and select the field containing the segmentation key for the "L" input.
Select Candidate flag and select the field containing the flag "Y" for selected candidates and "N" for rejected candidates. If this field does not exists on input, Data Management will create it on output as CANDIDATE_FLAG.
Optionally, select Remove non-candidates to remove records with Candidate flag set to "N".
Optionally, go to the Execution tab, and then set Web service options.