摘要
随着计算机技术和网络技术的的发展,各行各业积累的数据量越来越大。而专利信息集是目前世界上最大的技术信息集,几乎囊括了一切应用领域内的技术成果。为了提取隐含在其中的、人们事先不知道但又潜在有用的知识,将数据挖掘技术应用于专利信息分析,如采用聚类算法对专利文本进行挖掘、采用关联规则对专利发明人进行挖掘,以发现用户感兴趣的知识,并使之转化为有效的竞争情报。
With the development of computer and of life has been growing. Patent information set is Internet technique, the amount of data in every walk the world＇s largest set of technical information which practically includes technological achievements about all application areas. In order to get knowledge which is connotative, unknown and useful from practical data which is substantial, incomplete, noise, ambiguous and stochastic, data mining is applied into the analysis of patent information. For example, using clustering algorithms mine the text of patent and association rules mine the inventors of patent which can find the interested knowledge and make them into effective competitive intellizence.
出处
《情报科学》
CSSCI
北大核心
2008年第11期1672-1675,共4页
Information Science
关键词
数据挖掘
专利
聚类算法
关联规则
dada mining
patent
clustering algorithms
association rules
