一种改进的聚类和孤立点检测算法  

An Improved Algorithm for Clustering and Outlier Detection

作　　者：王永乐[1] 王希军[1] 

机构地区：[1]许昌职业技术学院信息工程系,河南许昌461000

出　　处：《许昌学院学报》2014年第5期24-27,共4页Journal of Xuchang University

摘　　要：对基于距离的聚类及基于密度的孤立点检测方法进行了分析研究,提出了一种基于距离和密度的聚类和孤立点检测算法DDBCOD.该算法根据距离和密度阈值对数据进行聚类,并发现数据中的孤立点.实验表明,该算法能够识别任意形状的聚类,对高维数据有效,能够很好的识别出孤立点.In this paper, a study is conducted on clustering and outlier detection, and a distance ＆ densitybased clustering and outlier detection algorithm（DDBCOD） is proposed. This algorithm records the datum points by distance and density threshold, and identifies outliers by density threshold. As is shown in the experimental results, the DDBCOD algorithm can cluster the dataset properly and discover clusters of arbitrary shapes. It is valid for high dimension dataset and it can find outliers accurately and validly.

关 键 词：聚类分析 孤立点 距离 密度 

分 类 号：G30[文化科学]