Total of 10,000 representative keywords are selected by feature selection algorithm,and then the vector space model(VSM) is built based on these keywords. The text ordered clustering is used to research of knowledge management subject finally.
本文以我国知识管理研究出现以来的5000余篇学术期刊载文为分析对象,以特征选择算法抽取出10,000个特征词为基础,采用向量空间模型(VSM)和文本挖掘技术中的有序聚类方法,揭示出知识管理学科领域的研究内容在我国经历的三个发展阶段,同时还从特征词入手对这三个发展阶段的研究特点进行了归纳总结。
参考来源 - 我国知识管理研究主题变化的计量分析·2,447,543篇论文数据,部分数据来源于NoteExpress
结果表明,根据有序聚类的原理确定粮食产量的状态,可使确定的分级标准更加合理;
The results showed that the classification standard was fixed with the sequential cluster principle.
并应用岩体风化差异系数计算各种方法下的各风化带的风化差异系数,综合分析,给出各最终推荐结果,按照有序聚类方法分析所得的结果进行风化岩体分带。
Then the difference coefficients of rock mass are calculated. With the integrative analysis of all the results, it is recommended to adopt the result of the way of ordered clustering.
本文借助于两个样品之间的差异度和有序样品的差异序列两个概念,提出了有序样品聚类的差异序列法法。
In this paper, with the aid of the two concepts of diversity between two samples and diversity sequence of ordered samples, diversity sequence method is presented for clustering ordered samples.
应用推荐