RAGONITE: Iterative Retrieval on Induced Databases and Verbalized RDF for Conversational QA over KGs with RAG

TL;DR

This paper introduces RAGONITE, a novel conversational QA system over RDF knowledge graphs that combines iterative retrieval from a derived database and verbalized facts, improving over traditional SPARQL-based methods.

Contribution

It proposes a new retrieval-augmented generation approach that fuses database query results and text search over verbalized KG facts with iterative retrieval capabilities.

Findings

Outperforms baseline methods on BMW knowledge graph

Supports iterative retrieval for improved answers

Effectively combines database and text search results

Abstract

Conversational question answering (ConvQA) is a convenient means of searching over RDF knowledge graphs (KGs), where a prevalent approach is to translate natural language questions to SPARQL queries. However, SPARQL has certain shortcomings: (i) it is brittle for complex intents and conversational questions, and (ii) it is not suitable for more abstract needs. Instead, we propose a novel two-pronged system where we fuse: (i) SQL-query results over a database automatically derived from the KG, and (ii) text-search results over verbalizations of KG facts. Our pipeline supports iterative retrieval: when the results of any branch are found to be unsatisfactory, the system can automatically opt for further rounds. We put everything together in a retrieval augmented generation (RAG) setup, where an LLM generates a coherent response from accumulated search results. We demonstrate the superiority of our proposed system over several baselines on a knowledge graph of BMW automobiles.