一种基于划分的孤立点检测算法-软件学报.PDFVIP

ISSN 1000-9825, CODEN RUXUEW E-mail: jos@ Journal of Software, Vol.17, No.5, May 2006, pp .1009−1016 http://www.j DOI: 10. 1360/jos171009 Tel/Fax : +86-10 © 2006 by Journal of Software. All rights reserved. ∗ 一种基于划分的孤立点检测算法 1,2 1+ 1 1 1 孙焕良 , 鲍玉斌 , 于 戈 , 赵法信 , 王大玲 1(东北大学 信息科学与工程学院,辽宁 沈阳 110006) 2(沈阳建筑大学 信息与控制工程学院,辽宁 沈阳 110015) An Algorithm Based on Partition for Outlier Detection 1,2 1+ 1 1 1 SUN Huan-Liang , BAO Yu-Bin , YU Ge , ZHAO Fa-Xin , WANG Da-Ling 1(School of Information Science and Engineering, Northeastern University, Shenyang 110006, China) 2(School of Information and Control Engineering, Shenyang Jianzhu University, Shenyang 110015, China) + Corresponding author: Phn: +86-24 Fax: +86-24 E-mail: baoyb@, Sun HL, Bao YB, Yu G, Zhao FX, Wang DL. An algorithm based on partition for outlier detection. Journal of Software, 2006,17(5):1009−1016. /1000-9825/17/1009.htm Abstract: Outliers are objects that do not comply with the general behavior of the data. The method of partition divides data space into a set of non-overlapping rectangular cells by partitioning every dimension into equal length. Statistical information of cells is used to find knowledge in datasets. There exists very large data skew in real-life datasets, so partition will produce many empty cells, which affects the efficiency of the algorithms. An efficient index structure called CD-Tree (cell dimension tree) is designed for indexing cells. Moreover, to guide partition and to opti