Turn documents and other digital content into useful, structured information. Intro to Information Extraction, Part 6, introduces the AI concepts behind extracting text, fields, and values, and explains how Microsoft Azure and Microsoft Foundry support these tasks. This course is Part 6 of 8 in the complete Microsoft AI Fundamentals AI-901 class. It builds on the earlier parts of the series, including Computer Vision, and focuses on information extraction. After completing this part, continue to Part 7: Machine Learning AI, where you will explore machine learning concepts and continue building your AI fundamentals.
Begin with the difference between unstructured content and structured data. Learn how extraction identifies entities and relationships, maps values to fields, and produces outputs such as JSON or tables. You will examine the stages of an extraction process, from identifying source content to transforming, storing, and integrating the results. Explore optical character recognition (OCR) and how it converts text in images and scanned documents into machine-readable content. Then compare Azure Vision Image Analysis, Azure Document Intelligence, Azure Content Understanding, and Azure AI Search. Learn where each service fits, how prebuilt and custom document models differ, and how analyzers define the information to extract. The lessons also introduce multimodal extraction from audio and video, along with knowledge mining for finding information across an organization’s content.
Business examples include invoices, receipts, contracts, healthcare documentation, and shipping records. You will consider extraction quality through precision, recall, and F1 score, and examine an example Azure solution architecture. Responsible AI topics address privacy, fairness, transparency, and human oversight when handling extracted information.
