首页|期刊导航|新疆大学学报(自然科学版中英文)|融合知识图谱与向量数据库的二级检索及反馈优化医学问答模型

融合知识图谱与向量数据库的二级检索及反馈优化医学问答模型OA

Medical Q&A Model Integrating Knowledge Graph and Vector Database with Two-Stage Retrieval and Feedback Optimization

中文摘要英文摘要

随着AI的迅速发展,大语言模型在医疗领域的应用潜力日益凸显,但通用大模型在医学专业领域存在知识不足、易产生幻觉等问题.本文以构建疾病医疗问答助手为核心任务,提出一种基于二级检索与反馈迭代机制的医学智能问答系统HQ-LLM(HealthQuery Large Language Model).整合医学百科、公共卫生部门等数据,构建医学知识图谱和中文医学问答数据集(Chinese Medical Knowledge Graph and Question-Answering Dataset,CnMedKG-QA);通过LoRA技术实现ChatGLM3-6B模型的轻量化微调,单张32 GB vGPU即可支撑训练过程,兼顾成本控制与训练效率;针对传统RAG局限,借助paraphrase-multilingual-MiniLM-L12-v2模型进行文本向量化,构建高相关性、高检索效率的向量数据库,创新设计粗粒度筛选与细粒度推理融合的二级检索机制;建立涵盖专业术语准确率、逻辑连贯性等维度的医学专属置信度评估体系,当综合置信度低于阈值时,动态优化检索关键词、扩展知识图谱追溯层数或补充外部权威数据,实现"检索-生成-评估-再检索"反馈迭代机制.结果表明,HQ-LLM在BLEU-4、Rouge-L、F1 Score和Accuracy指标上分别达到20.47%、42.12%、50.26%和79.20%,显著优于Qwen-UMLS-7B-Instruct等现有医学模型.

With the rapid advancement of AI,the application potential of large language models(LLMs)in the medical field has become increasingly prominent.However,general-purpose LLMs still suffer from insufficient professional knowledge and a tendency to generate hallucinations in medical domains.Focusing on the core task of constructing a disease-oriented medical question-answering assistant,this paper proposes a medical intelligent question-answering system named HQ-LLM(HealthQuery Large Language Model)based on a two-stage retrieval and feedback iteration mechanism.Multi-source data from medical encyclopedias and public health authorities are integrated to construct a Chinese Medical Knowledge Graph and Question-Answering Dataset(CnMedKG-QA).The LoRA(Low-Rank Adaptation)technique is adopted to implement light‐weight fine-tuning of the ChatGLM3-6B model,which can be supported by a single 32 GB vGPU for the training process,bal‐ancing cost control and training efficiency.To address the limitations of traditional RAG(Retrieval-Augmented Generation)systems,the paraphrase-multilingual-MiniLM-L12-v2 model is utilized for text vectorization to build a vector database with high relevance and retrieval efficiency.An innovative two-stage retrieval mechanism combining coarse-grained filtering and fine-grained reasoning is designed,along with a"retrieval-generation-evaluation-re-retrieval"feedback iteration mechanism.A medical-specific confidence evaluation system covering dimensions such as professional terminology accuracy and logical coherence is established.When the comprehensive confidence is below the threshold,retrieval key words are dynamically op‐timized,the number of knowledge graph tracing layers is expanded,or external authoritative data is supplemented to continu‐ously improve retrieval quality.Experimental results show that the HQ-LLM achieves 20.47%,42.12%,50.26%and 79.20%in BLEU-4,Rouge-L,F1 Score and Accuracy,respectively,significantly outperforming existing medical models such as Qwen-UMLS-7B-Instruct.

张骏伟;黄江;王翔;蒋晓红

重庆交通大学 机电与车辆工程学院,重庆 400074重庆交通大学 信息科学与工程学院,重庆 400074重庆交通大学 机电与车辆工程学院,重庆 400074重庆交通大学 交通运输学院,重庆 400074

信息技术与安全科学

检索增强生成知识图谱微调二级检索反馈迭代

RAGknowledge graphfine-tuningtwo-stage retrievalfeedback iteration

《新疆大学学报(自然科学版中英文)》 2026 (5)

562-576,15

重庆市技术创新与应用发展专项重大项目"基于大模型的汽车网络安全测评技术服务平台研发"(CSTB2024TIAD-STX0027).

10.13568/j.cnki.651094.651316.2026.01.12.0002

评论