Exam AI-102 topic 4 question 5 discussion

Correct!

1. Object 2. File https://learn.microsoft.com/en-us/azure/search/knowledge-store-projections-examples#define-an-object-projection Object projections are JSON representations of the enrichment tree that can be sourced from any node. In comparison with table projections, object projections are simpler to define and are used when projecting whole documents. Object projections are limited to a single projection in a container and can't be sliced. https://learn.microsoft.com/en-us/azure/search/knowledge-store-projections-examples#define-a-file-projection File projections are always binary, normalized images, where normalization refers to potential resizing and rotation for use in skillset execution. File projections, similar to object projections, are created as blobs in Azure Storage, and contain binary data (as opposed to JSON).

JSON --> Object projection is CORRECT. This projection type is used when working with unstructured or semi-structured JSON data in Azure AI Search. It allows the system to project JSON data as objects, making it easier to index and search through unstructured data, such as JSON documents. scanned data --> File projection is CORRECT. When dealing with extracted text from scanned PDF documents, file projection is the appropriate projection type. This type is designed for handling and storing text extracted from files like PDFs, images, or other documents within the Azure AI Search knowledge store.

1. Object 2. File

1. Object 2. File

Correct answer. https://learn.microsoft.com/en-us/azure/search/knowledge-store-projections-examples

we are dealing with unstructured JSON data here. Due to the potential irregularities in the data it will probably not be feasible using table projection. Using object projection is simpler and sometimes the only choice depending on the nature of the data.

java script OBJECT notation (JSON)

Object File https://docs.microsoft.com/en-us/azure/search/knowledge-store-projection-overview#types-of-projections-and-usage

Answer 1: Object Projection: Used when you need the full JSON representation of your data and enrichments in one JSON document. As with table projections, only valid JSON objects can be projected as objects, and shaping can help you do that. Answer 2: File Projection: Used when you need to save normalized, binary image files, eg: Optical Character Recognition (OCR) extracts text from binary files Not Answer: Table Projection: Used for data that's best represented as rows and columns, or whenever you need granular representations of your data (for example, as data frames). Table projections allow you to define a schematized shape, using a Shaper skill or use inline shaping to specify columns and rows.

Types of projections and usage A knowledge store is a logical construction that's physically expressed as a loose collection of tables, JSON objects, or binary image files in Azure Storage. TYPES OF PROJECTIONS AND USAGE Projection Storage Usage Tables Azure Table Storage Used for data that's best represented as rows and columns, or whenever you need granular representations of your data (for example, as data frames). Table projections allow you to define a schematized shape, using a Shaper skill or use inline shaping to specify columns and rows. You can organize content into multiple tables based on familiar normalization principles. Tables that are in the same group are automatically related. Objects Azure Blob Storage Used when you need the full JSON representation of your data and enrichments in one JSON document. As with table projections, only valid JSON objects can be projected as objects, and shaping can help you do that. Files Azure Blob Storage Used when you need to save normalized, binary image files.