About the Data Quality Services

The legal entities template includes data quality services for these attributes:

• Company name
• Country code
• URL Standardization and Cleaning

The data quality process examines values for the template's attributes, and adds any resulting validated, standardized values to each record in new enrichment-specific attributes. The original values mapped from your source datasets remain present and unchanged. See the topics linked above for processing details and added attributes.

About the Clustering Model

The legal entities model groups company records as follows:

First, by trusted_id. Records with the same trusted_id are always clustered together. Records with different trusted_ids are never clustered together.

Records with null/empty trusted_id are clustered based on similarity, meaning that they may be clustered with records that have a trusted _id.

Then, by similarity. Records with null or empty trusted_ids are clustered based on similarities between values for these attributes:

• Company name and alternate company names
• Legal entity name and alternative legal entity names, provided by the Company enricher
• Full address
• Country
• Website

Note: Generic descriptions, rather than specific attribute names, are listed to represent both the standard schema and the attributes added by the enrichers and other data transformations.

Clustering Examples for Legal Entities

Examples of Source Records Clustered Together

Example 1: These records are clustered together because they have the same or highly similar values for company name and some address fields, but different street addresses, indicating that they most likely represent different sites for the same legal entity.

Example 2: These records are clustered together because they have the same trusted_id.

Examples of Source Records Not Clustered Together

Example 1: These records are not clustered together because of missing or dissimilar address and website information, and only moderately similar company names, indicating that they most likely represent different legal entities.

Example 2: These records are not clustered together because they have different values for trusted_id.