基于OLAP和聚类分析结合的美国专利挖掘系统  

Design and realization of US patent mining system based on combination of OLAP and clustering analysis

作　　者：吕琳[1] 朱东华[1] 刘玉琴[1] 

机构地区：[1]北京理工大学管理与经济学院,北京100081

出　　处：《计算机工程与应用》2007年第25期186-187,217,共3页Computer Engineering and Applications

基　　金：国家自然科学基金(the National Natural Science Foundation of China under Grant No.70031010); 985哲学社会科学创新基地建设研究论文之一。

摘　　要：以美国授权专利数据库为实例,对OLAP及聚类分析技术进行了深入而细致的探讨。针对它们的共通性和差异性,提出了两者结合的美国专利挖掘系统的设计与实现方案,并给出了可视化结果。在此基础上,构建了数据挖掘系统的通用框架。结果表明,将OLAP和数据深层挖掘技术紧密配合、协调使用将是数据挖掘发展的一个方向和趋势。For US issued patent databases as example,OLAP and clustering technologies are in-depth discussed.In view of the similarity and difference of OLAP and clustering,the design and realization scheme of US patent mining system based on the combination of them is presented and their visualization outcomes are displayed,Based on that,a general frame of data mining system is constructed.The results show that the combinative method of OLAP and deep-seated mining technologies will be a trend of data mining development.

关 键 词：OIAP 聚类 数据挖掘 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]