This folder contains samples that demonstrate how to extract structured data from documents using Azure AI services.
Note
All data extraction samples provide both an accuracy and confidence score for the extracted data. The accuracy score is calculated based on the similarity between the extracted data and the ground truth data. The confidence score can be calculated based on OCR analysis confidence and logprobs in Azure OpenAI responses.
| Sample | Description | Example Use Cases |
|---|---|---|
| Text-based Extraction with Azure AI Document Intelligence and Azure OpenAI GPT-4o | Use Azure AI Document Intelligence prebuilt-layout and Azure OpenAI GPT models to extract structured data from documents using text. |
Predominantly text-based documents such as invoices, receipts, and forms. |
| Text-based Extraction with Azure AI Document Intelligence and Microsoft Phi | Use Azure AI Document Intelligence prebuilt-layout and Microsoft's Phi models to extract structured data from documents using text. |
Predominantly text-based documents such as invoices, receipts, and forms. |
| Vision-based Extraction with Azure OpenAI GPT-4o | Use Azure OpenAI GPT-4o models to extract structured data from documents using vision capabilities. | Complex documents with a mix of text and images, including diagrams, signatures, selection marks, etc. such as reports and contracts. |
| Multi-Modal (Text and Vision) Extraction with Azure AI Document Intelligence and Azure OpenAI GPT-4o | Improve the accuracy and confidence in extracting structured data from documents by combining text and images with LLMs. | Any structured or unstructured document type. |