Create Amazon Bedrock Knowledge Base and Data Source using Guide > Introduction to Create Amazon Bedrock Knowledge Base and Data Source using Guide recipe
  

Introduction to Create Amazon Bedrock Knowledge Base and Data Source using Guide recipe

The Create Amazon Bedrock Knowledge Base and Data Source using Guide recipe is based on REST and SOAP APIs. Use the recipe to create a knowledge base and resources automatically when the file is uploaded. You can also use the recipe to ask questions and get answers to the user's questions based on the uploaded file's contents using the Amazon Bedrock Large Language Model (LLM).
When you run the guide, you can specify the knowledge base name to create a new knowledge base or knowledge base ID to use the existing knowledge base, email for notification, and attach files. Optionally, you can prompt questions related to the file content. After you run the guide, the files are uploaded to the Amazon S3 bucket specified in the connection.
The process then creates a knowledge base with the specified name and vector database parameters. It creates data source for the Amazon S3 bucket in the knowledge base.
The process then starts the ingestion job and generates an answer. If the user has provided a prompt, it generates an answer based on the file content.
The process then sends an email notification to confirm successful job creation with the knowledge base ID, data source IDs, and the LLM answer if the prompt was specified. The recipe does not send an email notification about the errors. You can track the errors at the process execution status in Application Integration Console.
You can upload files with a maximum size of 5MB in the following formats:
.txt, .md, .html, .doc, docx, .csv, .xls, .xlsx, .pdf