- Newest
- Most votes
- Most comments
You can generate metadata for tables in your AWS Glue Data Catalog database using Amazon SageMaker Unified Studio's AI capabilities. Here's how to enable automated business name generation for all tables in a database:
- Navigate to the Data tab for your project in SageMaker Unified Studio
- In the left navigation pane, choose Data sources
- Select the data source connected to your Glue Data Catalog database
- Go to the details tab and enable the AUTOMATED BUSINESS NAME GENERATION configuration
This will allow SageMaker Unified Studio to use Amazon Bedrock's large language models to automatically generate business names for your tables.
Additionally, you can programmatically generate business names when creating assets by enabling the businessNameGeneration flag under predictionConfiguration in the CreateAsset API payload.
SageMaker Unified Studio can also generate other types of metadata including:
- Business descriptions and summaries
- Glossary terms
These AI-generated recommendations help automate data discovery and cataloging while reducing manual documentation tasks. You can edit, accept, or reject the generated metadata suggestions as needed.
For more comprehensive metadata enrichment, you might also consider creating custom metadata forms in SageMaker Unified Studio to add structured business context to your assets.
Sources
Using machine learning and generative AI in Amazon SageMaker Unified Studio - Amazon SageMaker Unified Studio
Create a metadata form in Amazon SageMaker Unified Studio - Amazon SageMaker Unified Studio
