Apply Generative AI in Data Preparation in Oracle Analytics

Use the dataset editor to apply generative AI to your data in Oracle Analytics. For example, you might might classify reviews as positive, negative, or neutral, or generate summaries of customer feedback comments.

For information about limitations with using AI Functions, see Known Issues with AI Functions.
  1. On your home page, hover over a dataset that you own, click Actions, then select Open.

    If the dataset contains a single table, you'll see the transform editor. If the dataset contains multiple tables, you'll see the Join Diagram. In the Join Diagram, right-click a data source and click Open to display the transform editor.

  2. To generate text, or apply a semantic filter, click Add Preparation Step in the Data Panel to add a column and select AI Generate or AI Filter from the function picker.
    Description of ai-functions-9.png follows
    Description of the illustration ai-functions-9.png

    For example:
    AI_GENERATE(  'Text: ' || TICKET_TEXT,  'Return a standardized category: Billing, Technical, Account, or Other. Return only the
      category.',  'model'.'id' )
    For example:
    CASE  WHEN AI_FILTER(  'Is this ticket about billing or payment?',  'Ticket: ' || TICKET_TEXT,  'model'.'id'  )  THEN 'Y'  ELSE 'N' END

    Note:

    If the dropdown for the model argument doesn't show the model you want to use, check that your model is registered in the Shared Folders area in the catalog, and that the model owner has granted you Read-Only or higher permissions. See Make a Generative AI Model Available in Oracle Analytics.
  3. Save the dataset.
After you finish authoring content, either you (if you have Full Control of the model) or the Generative AI model owner must grant intended 'readers' at least Read-Only permission to access the content.