ShelterW/chinese_medical_ner
收藏资源简介:
该README文件介绍了多个医疗领域的命名实体识别(NER)数据集,包括医渡云结构化4K数据集、中药说明书实体识别、中文糖尿病科研文献实体关系数据集DiaKG、中文医疗信息处理评测准CBLUE以及医药领域知识图谱。这些数据集涵盖了疾病和诊断、检查、检验、手术、药物、解剖部位等多种实体类型,并提供了数据集规模、数据来源等详细信息。
The README file provides detailed descriptions of multiple medical-related Named Entity Recognition (NER) datasets. Firstly, it describes the Yidu Cloud Structured 4K dataset and the Named Entity Recognition dataset for Chinese Electronic Medical Records, listing six types of entities: diseases and diagnoses, examinations, tests, surgeries, drugs, and anatomical parts. Secondly, it introduces the Chinese Herbal Medicine Instruction Entity Recognition dataset, which includes 1,000 training sets and 13 types of entities. Then, it mentions the Chinese Diabetes Research Literature Entity Relationship Dataset DiaKG, sourced from 41 Chinese diabetes expert consensus articles, annotating 22,050 medical entities and 6,890 entity relationships. Additionally, it describes the Chinese Medical Information Processing Evaluation Benchmark CBLUE, including the Chinese Medical Named Entity Recognition Dataset CMeEE and the Intelligent Dialogue Diagnosis Dataset IMCS. Finally, it details the Pharmaceutical Domain Knowledge Graph, centered around diseases, with 44,000 entities and 300,000 entity relationships, and lists the Chinese meanings, quantities, and examples of various entities.




