Use Machine Learning to Enrich Business Data with Swagger UI

After performing step 1 of the tutorial Use Machine Learning to Extract Information from Documents with Swagger UI to access and authorize the SAP Document AI Swagger UI, you need to create a client.

When you create a service instance for SAP Document AI, a default client is automatically created. A client is used in most of the endpoints to distinguish and separate data.

Free tier and trial users can only create one client. To see your list of clients:

1. Expand the GET /clients endpoint.

2. Click Try it out.

3. Enter a maximum number of clients to be listed in the limit field.

4. Click Execute.

You should receive a response like the following:

Use the DELETE /clients endpoint to delete the default client.

1. Expand the DELETE /clients endpoint.

2. Click Try it out.

3. Enter in the payload field the client id (default) you want to delete.

4. Click Execute.

You should receive a response like the following:

Use the POST /clients endpoint to create your own client. The clientId value created here will be used in other SAP Document AI endpoints.

1. Expand the POST /clients endpoint.

2. Click Try it out.

3. Enter your clientId and clientName values in the payload field in the format you see in Examples for payload parameter (c_29 and client 29, for example).

4. Click Execute.

You should receive a response like the following:

You can repeat step 1 to see the clientId and clientName of the client you have just created.

CAUTION:

When using the free tier option for SAP Document AI or a trial account, be aware of the technical limits listed in Free Tier Option and Trial Account Technical Constraints.

Use the POST /data/jobs endpoint to add your own master data records to the database to enrich the information extracted from documents.

1. Expand the POST /data/jobs endpoint.

2. Click Try it out.

3. Define the data in the payload field, so that the system knows which extracted field (using, for example, supplier IDs from master data) should be enriched.

   JSON Copy

   {
      "value":[
         {
            "id":"BE0001",
            "name":"Sliced Invoices",
            "address1":"Suite 5A-1204 123 Somewhere Street Your City AZ 12345",
            "bankAccount":"DE32245443233324",
            "taxId":"DE123456788"
         },
         {
            "id":"BE0002",
            "name":"Sliced",
            "address1":"Suite 9A-1204 123 Somewhere Boulevard Your City AZ 32323",
            "bankAccount":"DE32245443233325",
            "taxId":"DE123456789"
         }
      ]
   }    

4. Choose the enrichment data type businessEntity.

5. Enter your clientId (created in the previous step).

6. When you choose the enrichment data type business entity, you have the option to choose a subtype (supplier, customer or companyCode). In this example, choose supplier.

7. Click Execute.

What just happened?

In this example, in the payload field, several master data records (name, ID and address, for example) from 2 different suppliers (Sliced Invoices and Sliced) are provided, so this additional information can be added to the document extracted fields prediction when the information matches.

You should receive a response like the following with status PENDING:

Copy the id from the Response body to see the result of the enrichment data status in the next step.

Use the GET /data/jobs/{id} endpoint to see the status of the uploaded enrichment data.

1. Expand the GET /data/jobs/{id} endpoint.

2. Click Try it out.

3. Enter the id received in the POST /data/jobs endpoint as the id.

4. Click Execute.

You should receive a response like the following with status SUCCESS:

What just happened?

The refreshedAt parameter tells when the enrichment data job was refreshed for the last time. When the response is null, it means that the enrichment data has not yet been refreshed. Enrichment data is refreshed automatically every 4 hours. It might take up to 4 hours until the enrichment data prediction is available in the response.

Set data activation to manual, instead of using the default automatic refresh of enrichment data that takes place every 4 hours.

If you have already performed this step in the previous tutorial: Use Machine Learning to Enrich Employee Data with Swagger UI, you can skip it now. Set it to done and move directly to step 7.

1. Expand the POST /configuration endpoint.

2. Click Try it out.

3. Enter the following in the payload field:

   JSON Copy

   {
     "value": {
       "manualDataActivation":"true"
     }
   }  

4. Click Execute.

You should receive a response like the following:

Create a data activation job record to see new or updated enrichment data in the extraction results. Only activated enrichment data will be added to the extraction results.

1. Expand the POST /data/activation endpoint.

2. Click Try it out.

3. Click Execute.

You should receive a response like the following:

If you have already used this endpoint recently, you should receive a response like the following:

Wait until next data activation is possible to perform this step once again before moving to step 8.

SAP Document AI uses a globally pretrained machine learning model that currently obtains better accuracy results with invoices and payment advices in the languages listed in Supported Languages and Countries. The team is working to support additional document types and languages in the near future.

When enrichment data has been uploaded and fits to a certain prediction it is added to the results from the GET /document/jobs/{id} endpoint. To have the enrichment data in the prediction, you need to have the following part in the query of the POST /document/jobs endpoint (it is usually already there by default):

"enrichment": {
    "sender": {
      "top": 5,
      "type": "businessEntity",
      "subtype": "supplier"
    },
    "employee": {
      "type": "employee"
    }
}

Do the following:

1. Expand the POST /document/jobs endpoint.

2. Click Try it out.

3. Right click Sample Invoice 1, then click Save link as to download locally the document file for this enrich business data example.

You can also upload to SAP Document AI and enrich any document file in PDF or single-page PNG and JPEG format that has content in headers and tables, such as an invoice. In this case, make sure the data you define in the payload field, in step 4, matches your document fields.

1. Upload the document file you want to enrich.

2. In options, enter the list of fields to be extracted from the uploaded file (documentNumber, taxId, purchaseOrderNumber, shippingAmount, netAmount, senderAddress, senderName, grossAmount, for example), the client id you created in step 3 (c_29, for example), the document type (invoice, for example), receivedDate (2020-02-17, for example), the enrichment data type businessEntity and subtype supplier.

   JSON Copy

   {
      "extraction":{
         "headerFields":[
            "documentNumber",
            "taxId",
            "purchaseOrderNumber",
            "shippingAmount",
            "netAmount",
            "senderAddress",
            "senderName",
            "grossAmount",
            "currencyCode",
            "receiverContact",
            "documentDate",
            "taxAmount",
            "taxRate",
            "receiverName",
            "receiverAddress"
         ],
         "lineItemFields":[
            "description",
            "netAmount",
            "quantity",
            "unitPrice",
            "materialNumber"
         ]
      },
      "clientId":"c_29",
      "documentType":"invoice",
      "receivedDate":"2020-02-17",
      "enrichment":{
         "sender":{
            "top":5,
            "type":"businessEntity",
            "subtype":"supplier"
         },
         "employee":{
            "type":"employee"
         }
      }
   }

   Show all 41 lines Show less

3. Click Execute.

This is how the request should look like:

And that’s how the response looks like:

Copy the id from the Response body to get enrichment data prediction in the next step.

When enrichment data has been uploaded and fits to a certain prediction it is added to the results from the GET /document/jobs/{id} endpoint.

Enrichment data is refreshed automatically every 4 hours. It might take up to 4 hours until the enrichment data prediction is available in the response. If the enrichment data prediction is NOT available in the response in your first try, perform again (some hours later) steps 6 and 7. Do not perform steps 9 and 10 before you see the enrichment data prediction in the response in step 7.

1. Expand the GET /document/jobs/{id} endpoint.

2. Click Try it out.

3. Set returnNullValues and extractedValues to true.

4. Enter the id received in the POST /document/jobs endpoint as the id.

5. Click Execute.

You should receive a response like the following:

What just happened?

In this example, in the response, one of the extracted fields is the sender name Sliced Invoices. This information is enriched with the supplier ID enrichment data created in step 4. The prediction suggests the supplier ID BE0001 (from sender name Sliced Invoices) with higher probability than the supplier ID BE0002 (from sender name Sliced).

You have now successfully used the business entity to get enrichment data predictions for the document you uploaded to SAP Document AI.

To see a list of the enrichment data entries you have created:

1. Expand the GET /data endpoint.

2. Click Try it out.

3. Choose the enrichment data type businessEntity.

4. Enter your clientId.

5. Choose the enrichment data subtype supplier.

6. Click Execute.

You should receive a response like the following:

To delete enrichment data which has been uploaded before:

1. Expand the DELETE /data endpoint.

2. Click Try it out.

3. Define the data in the payload field, so that the system knows which data entry (using, for example, the data entry ID) should be deleted.

   JSON Copy

   {
      "value":[
         {
            "id":"BE0001"
         },
         {
            "id":"BE0002"
         }
      ]
   }

4. Choose the enrichment data type businessEntity.

5. Enter your clientId.

6. Choose the enrichment data subtype supplier.

7. Click Execute.

You should receive a response like the following:

You can repeat step 2 and delete the client you created in step 3.

Congratulations, you have completed this tutorial.