You have an Azure Cognitive Search instance that indexes purchase orders by using Form Recognizer You need to analyze the extracted information by using Microsoft Power Bl. The solution must minimize development effort. What should you add to the indexer?

To analyze extracted information in Power BI with minimal development effort, the best approach is to use a knowledge store with table projections. A knowledge store persists AI-enriched content from a Cognitive Search indexing pipeline. Table projections specifically shape this enriched data into a tabular format and save it to Azure Table Storage. Power BI can then connect directly to Azure Table Storage, consuming the well-structured, flattened data with minimal transformation required within Power BI itself. This is the most direct and efficient path for this use case.

B. a projection group: A projection group is a logical container for one or more projections (table, object, file). You would add a table projection inside a projection group, making "table projection" the more specific and correct component to add.
C. an object projection: An object projection saves the enriched data as a hierarchical JSON document in Azure Blob Storage. While Power BI can consume this, it requires significant data shaping and transformation within Power BI to parse the JSON, which is more effort than using pre-structured tables.
D. a file projection: A file projection is used to save binary files, such as images extracted or normalized during the enrichment process. It does not store the extracted textual or structured data needed for analysis in Power BI.

1. Microsoft Docs - Knowledge store in Azure Cognitive Search: This document introduces the knowledge store and its components. It states, "Table projections are useful when the data is dense, and you want to project it without loss of fidelity into rows and columns. Power BI dashboards and other apps that consume tabular data work well with this data structure."

Source: Microsoft, "Knowledge store in Azure Cognitive Search", Learn.microsoft.com. Accessed May 2024. Section: "Projections".

2. Microsoft Docs - Create a knowledge store with the REST API: This guide provides details on defining projections. It explicitly contrasts the use cases: "Choose table projections for tabular reporting and analysis in tools like Power BI."

Source: Microsoft, "Create a knowledge store with the REST API", Learn.microsoft.com. Accessed May 2024. Section: "Choose projection types".

3. Microsoft Docs - Example: Use Power BI with a knowledge store: This official tutorial provides a step-by-step guide for the exact scenario in the question. The entire tutorial is based on creating and using table projections to feed data into Power BI for analysis.

Source: Microsoft, "Example: Use Power BI with a knowledge store", Learn.microsoft.com. Accessed May 2024. Section: "Create a knowledge store", Step 3: "Add a knowledge store definition".