Demonstrate how to create an automated and Intelligent Inventory Manager solution capable of ingesting data in various formats (documents or pictures), extract objects, and validate product availability in a Datawarehouse.
- Customers will inquire about specific product names, technical characteristics, or descriptions of the desired product.
- The products sold by the client include a wide variety of nutritional supplements.
- The system should ingest data in various formats (.jpg, .png, .docx, .pdf, and others) and extract the content related to the product request.
- Based on the information provided by the customer, the system must query the database to verify the product's availability.
- If the product is not available, the system should suggest alternative products based on the database inventory.
- The solution must be fully automated without human intervention.
Repository folder structure:
-
images-aisearch: Pictures of nutrional supplements previously loaded into Azure AI Search. Each picture represents a document inside the Inventory Manager Azure AI Search Database.
-
images-lab-tests: Pictures used during the laboratories to query the existence of products (documents) inside Azure AI Search.
-
labs: Demos executed during the workshop.
We have created a Course Setup section to help you with setting up your development environment to execute the code and the labs.
The current solution has been built using the following services.
The Azure AI Search Service has been enrich using the following skillsets to extract information from multiple images.
