Unduplicate match

[cristty]

Hi,

I'm using a Dependent Unduplicate Match. I haven't found anywhere described the process of choosing the rows that go on the MATCH flow, vs the ones that go on the DUPLICATE flow. How does QS choose between them; how are the ones that go on the MATCH flow better than the ones from the DUPLICATE flow?

Thanks

[JRodriguez]

The key factor in the match process is the composite weight for the record. First you set a cut off value

The unduplicate match process set all records with composite weights above the match cutoff as a group of duplicates. Within the group the record with the highest composite weight and that matches to itself is declare the master record


IBM Infosphere QualityStage User guide, page 85

[cristty]

Thanks a lot