Ask Safely: Privacy-Aware LLM Query Generation for Knowledge Graphs

TL;DR

This paper introduces a privacy-aware method for generating knowledge graph queries using LLMs, which identifies and omits sensitive data to protect privacy without sacrificing query quality.

Contribution

It presents a novel approach that detects sensitive information in KGs and filters it out before LLM query translation, enhancing privacy in KG querying.

Findings

Preserves query quality while protecting sensitive data.

Effectively identifies sensitive information based on graph structure.

Maintains high accuracy in query generation with privacy safeguards.

Abstract

Large Language Models (LLMs) are increasingly used to query knowledge graphs (KGs) due to their strong semantic understanding and extrapolation capabilities compared to traditional approaches. However, these methods cannot be applied when the KG contains sensitive data and the user lacks the resources to deploy a local generative LLM. To address this issue, we propose a privacy-aware query generation approach for KGs. Our method identifies sensitive information in the graph based on its structure and omits such values before requesting the LLM to translate natural language questions into Cypher queries. Experimental results show that our approach preserves the quality of the generated queries while preventing sensitive data from being transmitted to third-party services.