Forum Stats

  • 3,782,418 Users
  • 2,254,643 Discussions


Filterting Candidate Records based on a value in Driver Record

Kavin T
Kavin T Member Posts: 29
edited Apr 18, 2018 10:52AM in Data Quality


using EDQ-CDS package.

I am looking at filtering the list of candidates that is sent into Matching process based on the value of an identifier in Driving record.

I was thinking of storing the value of identifier from driver record into Reference Data and use that to compare it against the candidate records. But unfortunately i am not able to store the value into Reference data in a Entity Match webservice.

Is there any other way that the identifer value can be stored into and used later in the process to filter candidate records? Thanks.





  • Mike-Matthews-Oracle
    Mike-Matthews-Oracle Member Posts: 1,544 Employee
    edited Apr 17, 2018 5:08AM

    Real-time matching services in CDS do not use the Reference Data port as this is designed for slow changing data. So in your case candidate records should just be records where the 'candidate' attribute is set to 1, rather than 0 for a driving record. You can use a check processor to filter them out easily before they go through to match.

  • Kavin T
    Kavin T Member Posts: 29
    edited Apr 17, 2018 8:43PM

    Thanks Mike for your response.

    I understand that candidate and driver record can be differentiated using 'candidate' attribute value. But my issue here is-

    I Would like to take an attribute value from the driver record store it in a attribute(which i was considering to use Ref Data, as it gets cleared every time it is loaded as part of writer). Use that attribute in the next process to filter the records on a condition that are fed into 'Match Prepare- Real-time' process(as in below snippet).


    Looks like i am not able to write the results into reference data. Is there any other alternative way of approaching this requirement?

    Note -

    I understand that this can taken into match process and filtered in the Match Rules. But that would lead me into duplicate rules for 2 different group of data.

    I am trying to filter it so only the right data is sent into matching process and the no Match rules is lesser and is ease of maintenance.(Less rules will bring better performance as well i believe)

  • Mike-Matthews-Oracle
    Mike-Matthews-Oracle Member Posts: 1,544 Employee
    edited Apr 18, 2018 10:51AM

    This would probably be possible (using an attribute of a driving record to filter out candidates) using a Message Handling Script. But I think it is probably the wrong approach to the problem. It would likely prove much more transparent just to include a In List/Is Value comparison on a match rule to meet the requirement.

  • Mike-Matthews-Oracle
    Mike-Matthews-Oracle Member Posts: 1,544 Employee
    edited Apr 18, 2018 10:52AM

    I am not sure why it would lead to duplicate rules for different data sets - can you explain? If you make the rule dependent on the value of an attribute that may or may not be populated, it can simply apply to one data set and not another in the same way as the filtering logic.

This discussion has been closed.