We improved all data product templates to ensure unique values for the tamr_record_id attribute, and made additional improvements to address enrichment and clustering in the Patient Mastering template. These changes are available only in newly created new data products.

Improved tamr_record_id Values

All data product mastering flows include a step to create a new primary key field tamr_record_id for each source record, ensuring that each source record has a unique primary key across all source datasets. Previously, this step created the tamr_record_id by concatenating the source dataset name and the source primary key. To further ensure that the tamr_record_id is unique, the step now creates the tamr_record_id by generating a 128-bit hashed value of the source dataset name and the source primary key.

Patient Mastering Template Updates

In addition to the tamr_record_id change above:

• This template now includes the new version of the address enrichment service, called Address Standardization, Validation, and Geocoding. This service provides powerful, accurate address matching, and returns a Google Place ID for matching addresses as well as location geocoding.

• Users now have more control over which records are clustered together. This template now includes a unified trusted_id attribute, to which you can map source columns that contain a unique record identifier, such as a Social Security Number or other internal identification number.

  The model always clusters together records with the same trusted_id values, and does not cluster together records with different trusted_id values. Records with a null/empty trusted_id are clustered with similar records as usual by the model, meaning that they may be clustered with records that have a trusted_id.

Learn more about this template.