Combine Elasticsearch's search relevance (BM25 + Semantic Search) with OpenAI's ChatGPT's question-answering capabilities to query your private/proprietary data.
Despite the incredible potential of ChatGPT, there are certain limitations like -
- Knowledge cutoff date
- Lacks access to relevant information
- Lack of knowledge about domain-specific content
In this hands-on workshop, We will learn how to connect ChatGPT to proprietary data stores using Elasticsearch and build question/answer capabilities for your data. In a demo, We will see how you can quickly convert your website, FAQ, or any documentation in prompt chat where your user can directly ask a question on your data. The entire interface will be built on python.
1. You have used ChatGPT :)
2. Good to have understanding around Elasticsearch (Not mandatory, Introduction will be cover)
Abhishek will give an overview of Elastic components. Hands-on session will be covered -
a) Creating an elastic cluster on the elastic cloud.
b) Performing CRUD operation on Elasticsearch.
c) Different types of search query.
GenAI in Elastic & ChatGPT + Elasticsearch by Ashish Tiwari
In this session, we will try to build question-answer functionality on your private data set (Same as chatGPT). We will see how you can leverage LLM and vector search to perform a semantic search on your data and with the help of hybrid search, you can retrieve the most relevant results. In this process, we going to connect with OpenAI API for chatCompletion. We will end up converting the sample website's data (Or any self-proprietary website) to prompt chat where you can direct ask a question on your data.