From the course: Building Applications Using Amazon Bedrock

Unlock the full course today

Join today to access over 24,200 courses taught by industry experts.

Conversation history

Conversation history - Amazon Bedrock Tutorial

From the course: Building Applications Using Amazon Bedrock

Start my 1-month free trial Buy for my team

Conversation history

“

- [Presenter] One of the limitations of our application in its current form is that you cannot ask the chatbot a follow-up question based on a previous answer. This is an extremely important feature of any chatbot application. In the natural flow of any conversation to give a better user experience, it helps with the chatbot was aware of how it answered previously and uses this as additional context to improve answers in the future. To help us with this process and to do some of the heavy lifting, we are going to introduce a new chain from long chain called conversational retrieval chain. When you provide this chain with a query, it will use documents that will retrieve from the retriever you specify as part of the typical RAG process, but then it goes a bit further. It'll also use the chat history to improve the quality of the RAG by taking into account past queries. This is really awesome and leads to a…

Contents