GET /fields

Returns a list of the available fields that your access token allows. This endpoint will return the field_id, the name of the field, the field description, and wether the field is a custom field or a Zuva built-in field. The a field_id can then be used to extract text from your documents using the POST /extraction endpoint.

Headers

Authorization (required) - Requires the standard authentication header used in the system.

Parameters

There are no parameters needed for this endpoint.

Code Sample
curl -X GET -H "Accept: application/json"  
-H "Authorization: Bearer <token>"
-H "Content-Type: application/json"  
-d "{}" https://<base_url>/api/v2/fields
fields, _ = zdai.fields.get()
Response Codes
  • 200 OK
    Request was successful.
  • 403 Forbidden
    Invalid or missing authentication token.
  • 500 Internal Server Error
    Something went wrong unrelated to the users request.

Response Data

  • field_id

    the field identifier. This can be used in other endpoints to extract text from documents

  • name

    the name of the field

  • description

    describes what the field is going to extract from your documents.

  • is_custom

    boolean to determine if the field is a custom field. If the field is a Zuva built-in field, false is returned, otherwise true.

  • bias

    the value or the relationship between precision and recall. When the bias setting is increased, Zuva will be over-inclusive, resulting in a higher recall rate. When the bias setting is decreased, Zuva will be under-inclusive, resulting in a higher precision rate. A Zuva built-in field will always return a value of zero

  • f_score

    balanced combination of Precision and Recall. This number is helpful as an overall measure of field accuracy. A Zuva built-in field will always return a value of zero

  • precision

    estimates the proportion of retrieved results that Zuva will find that are relevant (e.g., that the user would also highlight). The closer this number is to 1, the fewer highlights Zuva is predicted to find that the reviewer would not also find. A Zuva built-in field will always return a value of zero

  • recall

    estimates the proportion of results Zuva is expected to find out of all results the user would want Zuva to find. The closer this number is to 1, the more confident you can be that the field will not miss the language you are looking for. A Zuva built-in field will always return a value of zero

  • document_count

    the number of documents the field has been trained on. A Zuva built-in field will always return a value of zero

Response
[
    {
        "field_id": "<field_id>",
        "name": "Restoration of Rights and Remedies — Bond Indenture",
        "description": "This smart field captures ... ",
        "bias": 0,
        "f_score": 0,
        "precision": 0,
        "recall": 0,
        "document_count": 0,
        "is_custom": false
    }
]

Need Help?

Couldn’t find the information you were looking for or need more assistance?

Contact us at support@zuva.ai