中医药文献语义标注系统研究与开发

目的研究和开发支持中医和现代生物医学本体和术语集的语义标注系统。方法以MedPortal本体库和中医临床术语集等为术语资源库,设计语义标注系统工作流程和功能框架,并开发Web应用系统。结果构建了一个基于Web的中医药文献语义标注系统,支持语料库管理与维护、术语词典管理、语义标注和语义检索等功能,既可以为基于机器学习的信息抽取算法研究提供训练集,又能实现语义层面的多来源数据集成与知识融合。结论该中医药文献语义标注系统设计方案已经过实际项目验证,可为其他同类系统研发提供参考。

Research and Development of Semantic Annotation System for TCM Literature

Objective To research and develop a semantic annotation system supporting ontology and terminology of TCM and modern biomedicine. Methods Taking MedPortal ontology repository and TCM clinical terminology system as term resources, the workflow and functional framework of semantic annotation system were designed and a web application system was developed. Results A web-based semantic annotation system was built, supporting corpus management and maintenance, term dictionary management, semantic annotation and semantic retrieval, etc., which could not only provide training set for the research of information extraction algorithm based on machine learning, but also realize the multi-source data integration and knowledge fusion at the semantic level. Conclusion The design scheme of the semantic annotation system of TCM literature has been verified by actual projects, and can provide a reference for the development of other similar systems.