计算机科学 ›› 2016, Vol. 43 ›› Issue (8): 258-261.doi: 10.11896/j.issn.1002-137X.2016.08.052
左进,陈泽茂
ZUO Jin and CHEN Ze-mao
摘要： 通过改进传统K-means算法的初始聚类中心随机选取过程,提出了一种基于改进K均值聚类的异常检测算法。在选择初始聚类中心时,首先计算所有数据点的紧密性,排除离群点区域,在数据紧密的地方均匀选择K个初始中心,避免了随机性选择容易导致局部最优的缺陷。通过优化选取过程,使得算法在迭代前更加接近真实的聚类类簇中心,减少了迭代次数,提高了聚类质量和异常检测率。实验表明,改进算法在聚类性能和异常检测方面都明显优于原算法。
| [1] Yang Yu-zhou.Research and implementation of the clustering anomaly detection technology based on feature extraction[D].Chengdu:University of Electronic Science and Technology of China,2012(in Chinese) 杨宇舟.基于特征提取的聚类异常检测技术的研究与实现[D].成都:电子科技大学,2012 [2] Sun Na,Guo Yan-feng,Yao Yuan.Network data stream abnormal detection model based on SVM incremental learning method[J].Computer Engineering and Applications,2012,48(29):78-81(in Chinese) 孙娜,郭延锋,姚远.增量式SVM的数据流异常检测模型[J].计算机工程与应用,2012,48(29):78-81 [3] Luo Yong-jian.Research on Data Flow Anomaly Detection Algorithm Cluster-based[D].Harbin:Harbin Engineering University,2010(in Chinese) 骆永健.基于聚类的数据流异常检测算法的研究[D].哈尔滨:哈尔滨工程大学,2010 [4] Fu Ying-ding,Lan Ju-long.Kernel-based adaptation for affinity propagation clustering algorithm[J].Application Research of Computers,2012,29(5):1644-1650(in Chinese) 付迎丁,兰巨龙.基于核自适应的近邻传播聚类算法[J].计算机应用研究,2012,29(5):1644-1650 [5] Jiang Min,Pi De-chang,Sun Lan.Research on Density Clustering Algorithm with a Multiple Constraints[J].Computer Scie-nce,2011,38(10A):143-164(in Chinese) 江敏,皮德常,孙兰.一种多约束的密度聚类算法的研究[J].计算机科学,2011,38(10A):143-164 [6] Celeb M,Kingravi H,Vela P.A Comparative Study of Efficient Initialization Methods for the K-methods for the K-Means Clustering Algorithm [J].Expert Systems with Applications,2013,40(1):200-210 [7] Tzortzis G,Likas A.The minmax k-means clustering algorithm[J].Pattern Recognition,2011,44(4):866-876 [8] Jiang Da-yu.A fast and efficient parallel bisecting K-Means algorithm[D].Harbin:Harbin Engineering University,2013(in Chinese) 蒋大宇.快速有效的并行二分K均值算法[D].哈尔滨:哈尔滨工程大学,2013 [9] Zhu Jian-yu.Research and Application of K-means algorithm[D].Dalian:Dalian University of Technology,2013(in Chinese) 朱建宇.K均值算法研究及其应用[D].大连:大连理工大学,2013 [10] Han Zui-jiao.An Adaptive K-means initialization method based on data density[J].Computer Applications and Software,2014,1(2):182-187(in Chinese) 韩最蛟.基于数据密集性的自适应K均值初始化方法[J].计算机应用与软件,2014,31(2):182-187 [11] Macqueen J.Some methods for classification and analysis ofmultivariate observe[C]∥Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability.Berkeyey:University of California Press,1967:281-297 [12] Asuncion A,Newman D.UCI Machine Learning Respository[EB/OL].[2015-06-01].http://archive.ics.uci.edu/ml/datasets.html |
| No related articles found! |
|
||
首页