摘要
提出了一种基于知识网络的专家知识表示方法,这种方法利用维基百科作为背景知识,根据专家的科研成果和相关内容,自动构建专家个人的知识网络。在此基础上,分别从专家知识的广度和深度、研究兴趣划分、核心知识探测和知识稀缺性这些角度对专家知识网络进行了度量。最后,在一个真实科研组织中构建了专家的知识网络,并进行了度量和分析,验证了本文提出方法的有效性。
A graph-based knowledge representing method is proposed to surmount these difficulties.The knowledge network is constructed for every expert by leveraging the links structure of the Wikipedia to calculate the relatedness among the knowledge nodes extracted from the expert＇s publications.Thus the experts＇ knowledge can be measured precisely and directly.Furthermore,four metrics are presented to measure the experts＇ knowledge network from different aspects.Finally,the experiments on a real academic dataset demonstrate this approach is effective.
出处
《科学学研究》
CSSCI
北大核心
2010年第10期1521-1529,共9页
Studies in Science of Science
基金
国家自然科学基金课题(70671007)
教育部博士点基金课题(200800060005)
关键词
知识网络
知识管理
知识表示
知识度量
knowledge network
knowledge management
knowledge representation
knowledge measurement
