医学事件知识图谱数据集

本数据集是由我司构建的一个医疗文本数据集，可用于大模型预训练、监督微调、检索增强等场景。该数据集汇聚了丰富的临床数据，通过知识抽取技术，系统化整理疾病、药物、症状、治疗等医学实体及其复杂关系，为医疗研究、智能诊断、辅助决策提供坚实的数据支撑，加速医学领域的创新应用。数据集规模达到30万个实体，60万条关系，涵盖120万次就诊数据。

This medical text dataset is developed by our company, and is applicable to scenarios including LLM pre-training, supervised fine-tuning and retrieval-augmented applications. It aggregates abundant clinical data, and systematically organizes medical entities such as diseases, medications, symptoms, treatments and their complex relationships through knowledge extraction techniques. This dataset provides solid data support for medical research, intelligent diagnosis and auxiliary decision-making, and accelerates innovative applications in the medical field. The dataset consists of 300,000 entities, 600,000 relational pairs, and covers 1.2 million medical visit records.

搜集汇总

数据集介绍

特点

医学事件知识图谱数据集是一个由北方健康医疗大数据科技有限公司构建的医疗文本数据集，包含30万个实体和60万条关系，涵盖疾病、药物、症状、治疗等医学实体及其复杂关系。该数据集可用于大模型预训练、监督微调、检索增强等场景，支持医疗研究、智能诊断、辅助决策等多个应用。