Automatic PAM clustering algorithm for outlier detection

被引：9

作者：

Lei, Dajiang ^{[1
]}

Zhu, Qingsheng ^{[1
]}

Chen, Jun ^{[1
]}

Lin, Hai ^{[1
]}

Yang, Peng ^{[1
]}

机构：

[1] College of Computer, Chongqing University, Chongqing

来源：

Journal of Software | 2012年 / 7卷 / 05期

关键词：

Cluster validation; Outlier detection; PAM clustering algorithm; Subtractive clustering;

D O I：

10.4304/jsw.7.5.1045-1051

中图分类号：

学科分类号：

摘要：

In this paper, we propose an automatic PAM (Partition Around Medoids) clustering algorithm for outlier detection. The proposed methodology comprises two phases, clustering and finding outlying score. During clustering phase we automatically determine the number of clusters by combining PAM clustering algorithm and a specific cluster validation metric, which is vital to find a clustering solution that best fits the given data set, especially for PAM clustering algorithm. During finding outlier scores phase we decide outlying score of data instance corresponding to the cluster structure. Experiments on different datasets show that the proposed algorithm has higher detection rate go with lower false alarm rate comparing with the state of art outlier detection techniques, and it can be an effective solution for detecting outliers. © 2012 ACADEMY PUBLISHER.

引用

页码：1045 / 1051

页数：6