# Improve search results with Oracle Generative AI Agents, Vector Search, and OCI OpenSearch

- Source: https://docs.oracle.com/en/solutions/genai-vector-opensearch/index.html
- Date: 2024-12
- Type: reference-architecture
- Services: genai, opensearch, adb-s
- Tags: ai-ml, autonomous

## Summary (catalog)

Hybrid search combining vector similarity (ADB-S) with keyword search (OpenSearch). GenAI for query understanding and result re-ranking. Suitable for enterprise knowledge bases and document search.

## Architecture (fetched from source)

Deploy an AI-powered enterprise search engine using Oracle Generative AI Agents, Oracle Database 23ai Vector, and OCI OpenSearch 
 
 
 
 
- 
 
- 
 
 
 
 
 
- 
 
 
 
 
 
 
 
 
 
- 
 
 
 
 
 
 
 
 
 
- 
 
- 
 
- 
 
- 
 
- 
 
- 
 
- 
 
- 
 
- 
 
- 
 
- 
 
- 
 
 
 
 

 Previous 
 Next 
 JavaScript must be enabled to correctly display this content
 
 

 

 
 
 
- Improve search results with Oracle Generative AI Agents, Vector Search, and OCI OpenSearch 
 
- Deploy an AI-powered enterprise
 search engine using Oracle Generative AI Agents, Oracle Database 23ai Vector, and OCI OpenSearch 
 
 
 
 
Deploy an AI-powered enterprise
 search engine using Oracle Generative AI Agents, Oracle Database 23ai Vector, and OCI OpenSearch
 

 
 

 

 

 Search powered by artificial intelligence (AI) is intended to help
 employees in every line of business get immediate answers to complex queries. But
 writing programmatic queries using keywords or semantic search can be extremely
 challenging if not impossible for anyone without technical expertise. 
 
 

 
By using Oracle Generative AI services, anyone with a keyboard or voice
 command interface can:

 
 
- Quickly predict customer purchase behaviors by asking natural language questions 
 
- Instantly retrieve information from a knowledge base, and provide contextually
 relevant solutions to complex problems 
 
- Access and synthesize technical manuals and support forums to instantly retrieve
 step-by-step instructions 
 
 
By combining Oracle Generative AI Retrieval-Augmented Generation (RAG)
 Agents with a supported knowledge base, cloud engineers and enterprise architects can
 quickly:

 
 
- Build an intelligent search system that supports hybrid search (keyword
 search and semantic search), advanced data retrieval, and reranking to provide the
 most precise and relevant information 
 
- Provide a chat interface where any authorized user, with or without
 technical expertise, can use natural language to query enterprise data 
 
- Provide a managed vector data store and automated data ingestion
 pipeline to enable efficient storage and retrieval of complex data 
 
 

 
 
Replay the Webinar

 

 
Replay the webinar: 

 

 

 

 

 

 

 

 

 

 

 
 Title and Copyright Information 

 

 
Improve search results with Oracle Generative AI Agents, Vector Search, and OCI OpenSearch

 
G19700-01

 
December 2024

 
 Copyright © 2024, 

Oracle and/or its affiliates.