摘要
采用文本挖掘技术处理海量生物医学科技文献和文本注释型数据库 ,从而发现创新知识如基因、蛋白质、疾病、药物及其相互关系的研究是当前人工智能和数据挖掘领域研究的热点。本文对生物医学文献知识发现的研究内容、研究成果以及基于文本挖掘的关键技术诸方面进行了系统的分析和阐述。通过分析中医药学数据的特点 ,提出了基于文本挖掘的中医证候分子生物学知识发现研究 ,该方法的特点是综合利用中医药学文献和MEDLINE ,能够获得创新的证候与基因相关知识。初步实验表明 ,文本挖掘技术有望为证候的分子水平研究提供辅助和支撑手段。
Using text mining methods to process huge biomedical literature and text annotation databases for discovery of novel scientific knowledge such as gene, protein, disease, drug and the relationship among them, has been the foci in artificial intelligence and data mining research fields. The latest researches of biomedical literature knowledge discovery, including main issues, accomplishment, and the key methods from text mining perspective, are discussed. Furthermore, this paper proposes a text mining based approach to find the molecular level knowledge of Symptom Complex. This approach integrates the traditional Chinese medical literature and MEDLINE, and promises to discover novel knowledge about the relationship between Symptom Complex and genes. The preliminary experiments show that text mining would assist and support the molecular level study of Symptom Complex.
出处
《复杂系统与复杂性科学》
EI
CSCD
2004年第3期45-55,共11页
Complex Systems and Complexity Science
基金
国家科技部基础性工作项目 :中医药科技信息数据库的建设与共享 (2 0 0 2DEA3 0 0 42 )
关键词
生物医学文献
知识发现
数据库
数据挖掘
文本挖掘
中医药
text mining
knowledge discovery in literature
biomedicine
traditional chinese medicine
