高新企业竞争情报的聚类分析法  

The Cluster Analysis Method of Competitive Intelligence in Hi-tech Enterprises

作　　者：许淳熙[1] 

机构地区：[1]华中科技大学图书馆,武汉430074

出　　处：《图书情报工作》2007年第5期53-55,44,共4页Library and Information Service

摘　　要：企业竞争情报至为关键的一环是对所搜集并筛选后的信息情报进行分类。目前所采用的分类方法中由于一篇文献只划归为一类,使得文献的使用价值不能充分发挥。根据高新企业的实际情况,同一文献在不同的需求下,可以归为不同的类别。试图运用聚类分析法来对竞争情报进行分类,以有效地解决这一问题。高新企业竞争情报的聚类分析法的过程分为5个步骤,即建立竞争情报相似矩阵、将相似矩阵转换为等价矩阵、建立聚类标准、进行聚类计算、复制被聚类的文献并提交企业各使用部门。Classifying the filtered information and intelligence is a pivotal part of enterprises＇ competitive intelligence. The existing classification methods that an article only can be attributed to a category restricts the functions of competitive intelligence. The same article may belong to different classifications according to actual needs in Hi-tech enterprise. In order to solve this problem, the paper tries to use the cluster analysis method to deal with the competitive intelligence by five steps： set up similar matrix of documents, change the similar matrix into equivalent matrix, establish the standard of cluster analysis, calculate the results and submit to the various using departments of hi-tech enterprises.

关 键 词：竞争情报 高新企业 聚类分析 情报分类 等价关系 

分 类 号：F272[经济管理—企业管理]