InquiryAI Overview

InquiryAI is the first component of TigerGraph CoPilot.

It is a natural language query service that allows users to ask questions about their graph data in plain English. The service uses a Large Language Model (LLM) to convert the user’s question into a function call, which is then executed on the graph database. The service is designed to be easily extensible, allowing for the addition of new LLM providers and LangChain tools.

InquiryAI Architecture

InquiryAI Architecture

When a question is posed in natural language, CoPilot employs a novel three-phase interaction with both the TigerGraph database and a LLM of the user’s choice, to obtain accurate and relevant responses.

Phase 1

The first phase aligns the question with the particular data available in the database. CoPilot uses the LLM to compare the question with the graph’s schema and replace entities in the question by graph elements.

For example, if there is a vertex type of BareMetalNode and the user asks:

“How many servers are there?”

The question will be translated to:

“How many BareMetalNode vertices are there?”.

Phase 2

In the second phase, CoPilot uses the LLM to compare the transformed question with a set of curated database queries and functions in order to select the best match.

Phase 3

In the third phase, CoPilot executes the identified query and returns the result in natural language along with the reasoning behind the actions.

Using pre-approved queries provides multiple benefits.

  1. It reduces the likelihood of hallucinations, because the meaning and behavior of each query has been validated.

  2. The system has the potential of predicting the execution resources needed to answer the question.

Next Steps

Next, learn about TigerGraph CoPilot’s SupportAI Overview.

Return to TigerGraph CoPilot Overview page for a different topic.