Power Platform Community Forum Thread Details

Hello,

I trying to extract data from documents and classify them according to pre-defined categories.

Generally speaking this works okay'ish. The main issue is that the GPT introduces categories on its own, which I don't want. As far as I can understand, it is not something I can solve by modifying the prompt. E.g. here's an example of someone with the same issue.

However, this should be solvable by using a structured output with enum looking something like this:

{
  "type": "object",
  "properties": {
    "category": {
      "type": "string",
      "enum": [
        "Technology",
        "Finance",
        "Health",
        "Education",
        "Entertainment",
        "Science",
        "Politics",
        "Sports"
      ],
      "description": "The category assigned to the document. Must be one of the predefined values."
    }
  },
  "required": ["category"]
}

The issue is that I get an 'Arrays of primitive types are not supported' error.

Does anyone know if there any workaround for this I want to use AI Builder? Or is the only way to use a model via Azure?

Categories:

AI Builder

A standard categorize text prompt should do what you are looking to accomplish, unless I am completely missing what you are attempting.

Classify the <input text> into one of the <input Categories> , or 'unclear'. Analyze the key themes and context of the provided text and classify it into the most appropriate category from the list of categories. Take into consideration any subcategories, industry-specific terminology, and the possibility of ambiguous classification. If the text does not clearly fit into a single category, pick the closest category or label it as 'unclear'. Your answer may only have the exact text of the closest category. Do not explain and do not use categories not listed.