In this video tutorial, you’ll learn how to create and configure a Knowledge Base in Oracle AI Data Platform (AIDP) to help AI agents retrieve information from enterprise documents. By ingesting and indexing content, you can enable agents to access relevant organizational knowledge and provide more informed responses to user questions.
In this tutorial, you’ll learn how to:
- Create a Knowledge Base in Oracle AI Data Platform to support semantic retrieval of enterprise documents.
- Configure embedding models, chunk size, and chunk overlap to optimize how documents are processed and indexed for semantic search.
- Select a data source from a standard or external volume containing organizational documents.
- Ingest documents to generate vector embeddings and index content for retrieval.
- Monitor ingestion jobs using the Job Runs tab to verify that document processing completes successfully.
- Connect Knowledge Bases to AI agents through Retrieval-Augmented Generation (RAG) tools to retrieve relevant information during conversations.
By creating a Knowledge Base and preparing enterprise documents for semantic search, Oracle AI Data Platform enables AI agents to retrieve relevant information from organizational content. This approach helps improve the relevance of AI-generated responses and supports more effective knowledge-driven workflows. Watch the full tutorial to learn how to configure a Knowledge Base and prepare your documents for AI-powered retrieval.
For more information about creating and configuring Knowledge Bases, adding data sources, and monitoring ingestion jobs, see the Oracle Help Center documentation on Knowledge Bases. To learn how AI agents retrieve information from indexed documents, explore About Agent Tools, including the RAG tool.
Continue developing your Oracle AI Data Platform skills with self-paced learning modules and additional training resources in the Oracle AI and Analytics Learning Hub. You can also explore the Video Tutorial Playlists page for more step-by-step tutorials organized by product and task.
