基于结构模型的知识发现技术  

Knowledge discovery based on structural modeling

作　　者：范玉妹[1] 单平[2] 艾冬梅[1] 张德政[2] 

机构地区：[1]北京科技大学应用科学学院,北京100083 [2]北京科技大学信息工程学院,北京100083

出　　处：《北京科技大学学报》2008年第7期832-836,共5页Journal of University of Science and Technology Beijing

基　　金：国家“十五”科技攻关项目(No.2004BA721A01H07);北京市自然科学基金资助项目(No.4062022)

摘　　要：将一种结构建模方法用于分析知识结构中概念之间的关系,并将其用于中医医案解析,有效地获取医案中隐含的深层次的辨证论治规律.医案实例分析结果表明,由结构分析所得到的结构模型,不仅可根据系统中少量的、零散的已知概念关系推导出其他的大多数未知的概念之间的关系,并可使得已有结构关系凸现出来.结构的图分析可展现症状的关联度和重要程度,分析结果得到了医案验证和专家肯定.A structure model was used to analyze the relations between concepts in knowledge structure and parse the medical records of traditional Chinese medicine to gain connotative rules of treatment based on syndrome differentiation effectively. The analytic results of medical record examples show that the structure model gained by structural analysis can not only discover many un-known relations by few scattered existing information, but also protrude existing structural relations. The picture analysis of structure displays the correlation degree and importance of symptoms, and the results have been validated by medical records and affirmed by experts.

关 键 词：系统分析 知识获取 结构建模 中医 

分 类 号：N945.1[自然科学总论—系统科学]