AI Solution

Content Recommendation with Generative AI and Vector Store in HeatWave MySQL

Introduction

Oracle HeatWave GenAI includes HeatWave Chat, which provides an out-of-the-box chatbot that lets you quickly build your own AI-powered chatbot.

With this feature, you can interact with unstructured data stored in HeatWave Vector Store and in object storage using natural language. The HeatWave Chat interface preserves the context of questions to help enable a human-like conversation with follow-up questions. HeatWave maintains context from the history of questions asked, citations in the source documents, and the prompt to the large language model (LLM). This lets you verify the source of answers generated by the LLM. The context is maintained in HeatWave and is available for all applications using HeatWave.

The integrated Lakehouse Navigator lets you see data available in MySQL Database and object storage. You can then easily load selected data in HeatWave Vector Store in a fully automated way, helping you obtain more accurate and contextually relevant answers. You can search the entire database or restrict your search to a folder, instructing the LLM to retrieve information from that specific source. Several LLMs can be selected via HeatWave Chat, either built-in or accessible via the OCI Generative AI service.

Demo

Demo: Oracle HeatWave GenAI (5:40)

Note: HeatWave MySQL's generative AI and vector store capabilities are in beta and not generally available yet. Deeper resources will be shared when the product is available.