Bulk Matching External Records

📘

Record and Cluster Matching

Incoming records are matched against existing records or clusters in a mastering project.

Records: Tamr compares each incoming record against existing records in the unified dataset to determine any matching records. For each record that finds a match, the response includes the matching record id pair and the confidence score.

Clusters: Tamr compares each incoming record against existing clusters to find any matching clusters. For each record that finds a match, the response includes the matching record id and cluster id pair, and the average confidence score.

Checklist before proceeding

• At least one mastering project exists (Creating a Project ).
• At least one dataset is added to the project and is schema mapped to the project's unified dataset (Adding a Dataset ).
• Generate Record Pairs and the Update Results jobs have both been executed.

Bulk Match and Published Clusters

If clusters have been published, bulk match will be performed against records and clusters from the most recent cluster publication. If clusters have never been published, bulk match will be performed against the current records and clusters.

Synchronous Operation

1. Bulk Match: POST /dedup/match/records/{name}
   Submit records in a stream for matching against existing records, where {name} is the name of the unified dataset of a Mastering project.

OR

1. Bulk Match: POST /dedup/match/clusters/{name}
   Submit records in a stream for matching against existing clusters, where {name} is the name of the unified dataset of a Mastering project.

Asynchronous Operation

1. Bulk Match: POST /projects/{project}:bulkMatch
   Submit records in a single batch for matching, where project is the name of Mastering project, and type is either of the keywords records or clusters. Additionally, capture the id of the submitted job from the response.
2. Wait For Job: GET /operations/{operationId}
   Poll the status of the job submitted in Step 1 using the captured {id} until status SUCCEEDED received.
3. Export Results: GET /projects/{project}/results/{operationId}
   Export the materialized bulk record match results dataset using the captured {id} from Step 1.