Adding a custom service for automatic term assignment (IBM Knowledge Catalog)

You can use your own customized service for automatic term assignment in metadata enrichment.

Base Premium Standard Unless otherwise noted, this information applies to all editions of IBM Knowledge Catalog.

Instead of using the built-in machine learning model, you can create and use your own ML models for generating term assignments and suggestions.

Prerequisites

Watson Machine Learning must be deployed in your IBM Software Hub environment. A custom service must be available.

To learn how you can use a model and scoring function deployed to Watson Machine Learning for automatic term assignment in Cloud Pak for Data, work with a sample notebook for custom term assignment based on a Watson Machine Learning model. You can download the sample notebook file from the IBM Knowledge Catalog samples GitHub repository: https://github.com/IBM/knowledge-catalog-samples/tree/main/metadata-enrichment/term-assignment/custom-term-assignment

Required permissions

To add a custom service for automatic term assignment, you must have the Admin role in the project to which the enrichment settings apply.

To add a custom service for automatic term assignment to a project:

1. Edit the project's default enrichment settings. Open an existing metadata enrichment asset and click Default settings. You can also go to the project's Manage page and select Tools > Metadata enrichment.
2. Select Custom service and click Select service.
3. Select a deployment space for the service and select the service deployment.
4. Optional: Test your service deployment.
5. Optional: Define input and output mappings by adding transformation code.
6. Review your setup.
7. To complete the setup, click Select.