Azure ML Studio Integration for AI Governance Models

You can integrate an instance of Azure Machine Learning (ML) Studio with OneTrust to enable automatic mapping of model attributes (name, version, creation date, creator, description, type of learning) from Azure ML studio into corresponding fields in OneTrust. This offers Data Scientists a workstream that is more efficient than manual data entry and an ability to maintain up-to-date model information.

It is achieved by setting up a webhook subscription in Azure Events to trigger a linked record to be created in OneTrust whenever a new object is registered as an "Asset" in Azure ML Studio.

Things to Know

  • If the "Type of Learning" is not present in OneTrust, the application will automatically create a new entry and populate it to the dropdown list.

  • The "Hosting Provider" field in OneTrust should automatically be set as 'Microsoft Azure' when a new model is created in Azure ML Studio.

To create the Integration

  1. On the Integration menu, select Gallery.

    The Integrations Gallery screen appears.

  2. Search for Azure Machine Learning and select the tile.

    The Gallery Details screen appears.


    Integrations gallery - Azure Machine Learning
  3. Click the Add button for Azure Machine Learning Model Registration.

    The Add Workflow Name modal appears.


    Azure Machine Learning Model Registration
  4. Complete the fields, then click the Create button.

    The Integration Builder screen appears.

    Note:

    You can select None for the Credential field.


    Integration builder - Add workflow name
  5. The Built-in Webhook Trigger action will capture the model's details using a number of variables in the provided sample body:
    Note:

    You will need to copy this webhook URL in order to configure the webhook in Azure. See Connecting Azure Events.


    Webhook action
  6. The POST Create an Object with Values Defined in the Request action will return the model's name, organization group ID, version, and description using our provided variables:

    POST Create an Object with Values Defined in the Request - Request details
  7. Save and activate the integration.

Connecting Azure Events

Required Configuration

Tip:

For more information on event subscriptions and model registration, see Microsoft's guide on trigger applications here.

  • Event Subscription with selected (optional) Event Types:

    • Model registered

    • Model deployed

    • Run completed

    • Dataset drift detected

    • Run status changed

  • Authenticate the webhook URL. For more information, see Microsoft's webhook event delivery guide here.

  • Endpoint Type as the specified webhook URL from the integration (see step 5 above) appended with a OneTrust API Key. For more information on creating and obtaining API Keys, see Managing OAuth 2.0 Keys.

    JSON 
    Example: https://app.onetrust.com/integrationmanager/api/v1/webhook/{API-KEY}