Knowledge Graph-Driven Retrieval-Augmented Generation: Integrating Deepseek-R1 with Weaviate for Advanced Chatbot Applications

TL;DR

This paper presents a retrieval-augmented generation framework that combines biomedical knowledge graphs with LLMs to produce more accurate, verifiable, and contextually relevant responses in medical chatbot applications.

Contribution

It introduces a novel integration of structured biomedical knowledge graphs with LLMs using retrieval-augmented generation for improved medical chatbot reliability.

Findings

Reduces hallucinations in generated responses

Enhances factual accuracy and clinical relevance

Improves response clarity in biomedical chatbot applications

Abstract

Large language models (LLMs) have significantly advanced the field of natural language generation. However, they frequently generate unverified outputs, which compromises their reliability in critical applications. In this study, we propose an innovative framework that combines structured biomedical knowledge with LLMs through a retrieval-augmented generation technique. Our system develops a thorough knowledge graph by identifying and refining causal relationships and named entities from medical abstracts related to age-related macular degeneration (AMD). Using a vector-based retrieval process and a locally deployed language model, our framework produces responses that are both contextually relevant and verifiable, with direct references to clinical evidence. Experimental results show that this method notably decreases hallucinations, enhances factual precision, and improves the clarity of generated responses, providing a robust solution for advanced biomedical chatbot applications.