The final, formatted version of the article will be published soon.
ORIGINAL RESEARCH article
Front. Public Health
Sec. Digital Public Health
Volume 13 - 2025 |
doi: 10.3389/fpubh.2025.1540946
This article is part of the Research Topic mHealth and smartphone apps in patient follow-up View all 18 articles
Building an Intelligent Diabetes Q&A System with Knowledge Graphs and Large Language Models
Provisionally accepted- 1 Southwest Jiaotong University, Chengdu, China
- 2 Guangxi Police College, nanning, China
- 3 Yangzhou University, Yangzhou, Jiangsu Province, China
- 4 Macao Polytechnic University, Macau, Macao, SAR China
This paper introduces an intelligent question-answering system that delivers personalized medical information to diabetic patients by integrating large language models with knowledge graphs. The system uniquely combines a Neo4j-based knowledge graph with the Baichuan2-13B and Qwen2.5-7B models, leveraging Low-Rank Adaptation (LoRA) and prompt-based learning to enhance semantic understanding and response quality. Experimental results demonstrate the system's high performance in entity recognition and intent classification, achieving precision scores of 85.91% and 88.55%, respectively. Additionally, the integration of a structured knowledge graph substantially enhances both the accuracy and clinical relevance of the system's responses.This study underscores the potential of integrating knowledge graphs with large language models for advancing diabetes management, offering an effective framework for personalized healthcare interventions.
Keywords: knowledge graph, Q&A system, Large language models, Prompt learning, personalized health management
Received: 06 Dec 2024; Accepted: 27 Jan 2025.
Copyright: © 2025 Qin, Wu, Zang, Chen, Zhang and Wong. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Zhenkai Qin, Southwest Jiaotong University, Chengdu, China
Dongze Wu, Guangxi Police College, nanning, China
Zhidong Zang, Yangzhou University, Yangzhou, 225009, Jiangsu Province, China
Xiaolong Chen, Macao Polytechnic University, Macau, Macao, SAR China
Hongfeng Zhang, Macao Polytechnic University, Macau, Macao, SAR China
Cora UnIn Wong, Macao Polytechnic University, Macau, Macao, SAR China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.