You can use a Duplicate Record Exception transformation to identify clusters of duplicate data that needs manual review. The match scores of records in clusters determines the potential duplicates. You can configure upper and lower thresholds for match scores in the transformation. The upper and lower thresholds define the degree of similarity.
A cluster contains related records that a matching operation groups together. The Match transformation creates clusters using the duplicate analysis operation and the identity resolution operation. Each record in a cluster has the same cluster ID. When the lowest match score in a cluster is between the upper and lower thresholds, the Duplicate Record Exception transformation identifies the cluster as a duplicate record exception cluster. The Match transformation adds a cluster ID value column to all the records. Duplicate records receive the same cluster ID.
The lowest record score in a cluster determines the cluster type. A cluster might have 11 records that have a match score of 0.95 and one record with match score of 0.79. If the upper threshold is 0.9 and the lower threshold is 0.8, the Exception transformation writes the records to the unique records table.
ASK INFAPreview
