Skip to content

Feedback Loop

What is a Feedback ID?

In the process of annotating documents, a unique identifier, known as a Feedback ID, is returned. The Feedback ID is generated for each annotated document, which can be used to gather specific feedback on the results provided. Feedback means that you are not sending us corrections, but the full truth.

Provide feedback

Users are encouraged to provide feedback for successfully annotated documents within a 24-hour window of the annotation. This timeframe is designed to capture timely insights and enhance the efficiency of our feedback loop. In exceptional cases requiring an extension of this limit, users can reach out to our support team for assistance.

To ensure a smooth feedback process, it's crucial to be aware of potential errors:

  • 404 Not Found Error: Attempting to provide feedback after the 24-hour window will result in a 404 Not Found error. Users should adhere to the recommended timeframe for optimal engagement.

  • 409 Conflict Error: Rapidly submitting multiple feedback calls for the same document may result in a 409 Conflict error for some calls. Subsequently, this may lead to 404 errors once the initial call is complete and temporary stored data is removed.

Example request

POST /v1/feedback:create

https://api.stag.ssn.visma.ai/v1/feedback:create

Authorization - Bearer Token

Token: demo

Body - raw (json)

Body 
{
    "id": "65226d1b-1c70-425c-9288-a29106928ef2",
    "tags": [
        "foo",
        "bar",
        "baz",
        "qux"
    ],
    "true_values": {
        "currency": "DKK",
        "invoice_number": "1202942",
        "order_date": {
            "day": 1,
            "month": 1,
            "year": 2019
        },
        "payment_due_date": {
            "day": 1,
            "month": 1,
            "year": 2019
        },
        "total_excl_vat": 225.0,
        "total_incl_vat": 300.0,
        "total_vat": 75.0
    }
}

Purpose of Feedback Data:

If you are sending us feedback, you're not just sending info, but you are helping us gather insights and train the model. NB: It is not our goal to change our process into human-in-loop ("human in loop" means integrating human employees into the machine learning pipeline, so that they can continuously validate e.g. documents and AI models).

Feedback data is used for:

  • Customer analysis and insights
    • Measure how Smartscan is performing and perceived by the end-user.
    • Understand what the customer’s goal is with particular fields.
    • The feedback data will be used to generate stats (dashboard) regarding the results provided by Smartscan and how end users interact with these results.
  • Improvement loop to train the model
    • We can use the Feedback IDs to look at particular documents and train the system on that. This is not an automated loop.

Deleting Feedback Data

Head over to our Data Deletion Policy.