A new Algorithm for Qutlier Detection based on Offset

摘要：

Outlier detection is a hot topic of data mining.After studying the existing classical algorithm of detecting Outliers, this paper proposes a new algorithm for outlierDetection based on offset, and makes a new definition for outlier; This detection algorithm is a method based on clustering analysis. It includes cluster modeling and data detecting outliers.The experiment results on IRIS show that this algorthm can detect outliers effectively.

关键词： Outlier Chustering Offset Mathematical ezpectation Strandard deviation

作者: Yue Zhang Jie Liu Bo Song

作者单位: Software College Shenyang Normal University Sheyang 110034,China Software College Shenyang Normal University Shenyang 110034,China

会议类型: 国际会议

会议名称: The Fifth International Conference on Information Assurance and Security(第五届信息保障与安全国际会议)

会议地点: 西安

会议语种:英文

页码: 3-6

在线出版日期: 2009-08-18（万方平台首次上网日期，不代表论文的发表时间）