Robust support vector data description for outlier detection with noise or uncertain data

Robust support vector data description for outlier detection with noise or uncertain data
复制标题

用于噪声或不确定数据的异常值检测的稳健支持向量数据描述

DOI:
10.1016/j.knosys.2015.09.025
复制
发表时间:
2015-12
影响因子:
8.8
通讯作者:
Li Fenglian
Li Fenglian
中科院分区:
计算机科学1区
文献类型:
--
作者:
Chen Guijun;Zhang Xueying;Wang Zizhong John;Li Fenglian

文献摘要

参考文献

被引文献

相似文献

作为一类分类方法的示例,支持向量数据描述(SVDD)提供了提高异常值检测性能并减少许多实际应用中异常值发生所造成的损失的机会。然而,由于异常值有限,SVDD模型仅使用正常数据构建。在这种情况下,当正常数据包含噪声或不确定性时,SVDD很容易导致过拟合。本文提出了两种新的SVDD方法,即R-SVDD和εNR-SVDD,它们是通过引入基于截止距离的每个数据样本的局部密度和带有负样本的ε不敏感损失函数来构造的。通过对 10 个 UCI 数据集的大量实验,我们证明了所提出的方法可以提高 SVDD 对于带有噪声或不确定性的数据的鲁棒性。实验结果表明,所提出的εNR-SVDD在检测率和误报率方面优于其他现有的异常值检测方法。同时,所提出的 R-SVDD 还可以仅使用正常数据实现更好的异常值检测性能。最后,所提出的方法成功地用于检测基于图像的传送带故障。
As an example of one-class classification methods, support vector data description (SVDD) offers an opportunity to improve the performance of outlier detection and reduce the loss caused by outlier occurrence in many real-world applications. However, due to limited outliers, the SVDD model is built only by using the normal data. In this situation, SVDD may easily lead to over fitting when the normal data contain noise or uncertainty. This paper presents two types of new SVDD methods, named R-SVDD andεNR-SVDD, which are constructed by introducing cutoff distance-based local density of each data sample and theε-insensitive loss function with negative samples. We have demonstrated that the proposed methods can improve the robustness of SVDD for data with noise or uncertainty by extensive experiments on ten UCI datasets. The experimental results have shown that the proposedεNR-SVDD is superior to other existing outlier detection methods in terms of the detection rate and the false alarm rate. Meanwhile, the proposed R-SVDD can also achieve a better outlier detection performance with only normal data. Finally, the proposed methods are successfully used to detect the image-based conveyor belt fault.
DOI: 10.1016/j.asoc.2010.12.012
发表时间: 2011-04
期刊: Appl. Soft Comput.
影响因子: --
作者:
Xiaoying Pan;L. Jiao
通讯作者: Xiaoying Pan;L. Jiao
DOI: 10.1023/b:mach.0000008084.60811.49
发表时间: 2004-01-01
期刊: MACHINE LEARNING
影响因子: 7.5
作者:
Tax, DMJ;Duin, RPW
通讯作者: Duin, RPW
DOI: --
发表时间: 1996
期刊: --
影响因子: --
作者:
C. Merz
通讯作者: C. Merz
DOI: 10.1162/089976603321891855
发表时间: 2003-07-01
期刊: NEURAL COMPUTATION
影响因子: 2.9
作者:
Keerthi, SS;Lin, CJ
通讯作者: Lin, CJ
DOI: 10.1016/j.patcog.2005.03.020
发表时间: 2005-10-01
影响因子: 8
作者:
Lee, K;Kim, DW;Lee, KH
通讯作者: Lee, KH