Enterprise Knowledge Graph field enrichment

Document AI uses Google's Knowledge Graph to normalize and enrich entity extraction (for supported fields). For example, the addresses "123 Main St Apt 1" and "123 Main street # 1" could be normalized to the same standardized address.

For each supported field, if a matching entity is found in the Knowledge Graph, then Document AI also returns a normalizedValue in addition to the extracted field.

Enriched fields

Here are the processors and fields that support entity enrichment:

Note: Enriched fields are subject to change with new processor versions. Extracted fields should remain consistent between versions. Please follow the Release Notes for Document AI updates.

Procurement

• Expense Parser
• supplier_address
• supplier_name
• supplier_phone

• Invoice Parser
• supplier_address
• supplier_name
• supplier_phone

Lending

This content is available to approved customers. You can view this content after you have been approved and added to the appropriate allowlist.

Sample response

The EKG enriched values can be found in the entities field under normalizedValue as shown in the following truncated sample:

    {
      entities: [
        {
          "textAnchor": {
            "textSegments": [ ... ],
            "content": "Google Singapore"
          },
          "type": "employer_name",
          "mentionText": "Google Singapore",
          "confidence": 0.69933707,
          "pageAnchor": {
            "pageRefs": [
              {
                "boundingPoly": {
                  "normalizedVertices": [ ... ]
                }
              }
            ]
          },
          "id": "9",
          "normalizedValue": {
            "text": "Google Asia Pacific, Singapore"
          }
        }
      ]
    }

In the sample, the original employer_name "Google Singapore" has been normalized to "Google Asia Pacific, Singapore".

In the Cloud console, the normalized fields are annotated with G. For example:

Sample normalized field shown in the web application.