详细信息
文献类型:期刊文献
中文题名:基于蚁群聚类算法的离群挖掘方法
英文题名:Research on Fault Diagnosis Based on Outlier Mining
作者:杨欣斌[1];孙京诰[1];黄道[1]
机构:[1]华东理工大学信息学院,上海200237
年份:2003
卷号:39
期号:9
起止页码:12
中文期刊名:计算机工程与应用
外文期刊名:Computer Engineering and Applications
收录:CSTPCD;;北大核心:【北大核心2000】;CSCD:【CSCD2011_2012】;
语种:中文
中文关键词:离群数据;数据挖掘;蚁群;聚类
外文关键词:outlier mining,data mining,ant colony,clustering
摘要:离群挖掘是数据挖掘研究的重要内容,在实际生活中获得广泛应用。该文首先给出了离群数据的量化定义,并用基于蚁群的聚类学习方法,产生了状态空间的整体特征。然后结合具体的设备对象,提出了离群数据的挖掘方法。最后进行了实验验证,结果表明该文提出的方法是有效的。
Outlier mining is an important issue in data mining community.It was applied widely in practice.A quantitative definition is proposed at the beginning of this paper.The integral character of the state space is created using the clustering method based on ant colony.And a method of mining outlier is put forward according to the practical device.At the end,an experiment has been done and the result proves the effectiveness of the method.
参考文献:
正在载入数据...
