面向大规模噪声数据的软性核凸包支持向量机

doi:10.3969/j.issn.0372-2112.2018.02.013

电子学报 ›› 2018, Vol. 46 ›› Issue (2): 347-357.DOI: 10.3969/j.issn.0372-2112.2018.02.013

面向大规模噪声数据的软性核凸包支持向量机

顾晓清^1,2, 倪彤光², 姜志彬¹, 王士同¹

1. 江南大学数字媒体学院, 江苏无锡 214122;
2. 常州大学信息科学与工程学院, 江苏常州 213164

收稿日期:2016-10-24 修回日期:2017-10-17 出版日期:2018-02-25
- 通讯作者:
- 倪彤光
- 作者简介:
- 顾晓清,女,1981年出生,江苏常州人,2017年获江南大学博士学位,现任常州大学讲师,研究方向为模式识别,模糊系统.E-mail:czxqgu@163.com;姜志彬,男,1991年出生,山东烟台人,江南大学博士研究生,主要研究方向为模式识别.;王士同,男,1964年出生,江苏扬州人,江南大学教授、博士生导师,主要从事人工智能、模式识别、模糊系统、医学图像处理和生物信息学等方面的研究工作.
- 基金资助:
- 国家自然科学基金 (No.61572236，No.61572085）; 江苏省自然科学基金 (No.BK20160187）

Soft Kernel Convex Hull Support Vector Machine for Large Scale Noisy Datasets

GU Xiao-qing^1,2, NI Tong-guang², JIANG Zhi-bin¹, WANG Shi-tong¹

1. School of Digital Media, Jiangnan University, Wuxi, Jiangsu 214122, China;
2. School of Information Science and Engineering, Changzhou University, Changzhou, Jiangsu 213164, China

Received:2016-10-24 Revised:2017-10-17 Online:2018-02-25 Published:2018-02-25
- Corresponding author:
- NI Tong-guang
- Supported by:
- National Natural Science Foundation of China (No.61572236, No.61572085); Natural Science Foundation of Jiangsu Province, China (No.BK20160187)

摘要/Abstract

摘要： 现有的面向大规模数据分类的支持向量机（support vector machine，SVM）对噪声样本敏感，针对这一问题，通过定义软性核凸包和引入pinball损失函数，提出了一种新的软性核凸包支持向量机（soft kernel convex hull support vector machine for large scale noisy datasets，SCH-SVM）.SCH-SVM首先定义了软性核凸包的概念，然后选择出能代表样本在核空间几何轮廓的软性核凸包向量，再将其对应的原始空间样本作为训练样本并基于pinball损失函数来寻找两类软性核凸包之间的最大分位数距离.相关理论和实验结果亦证明了所提分类器在训练时间，抗噪能力和支持向量数上的有效性.

关键词: 大规模数据, 噪声, 软性核凸包, pinball损失函数, 分类

Abstract: Current support vector machines (SVMs) for large-scale datasets classification problems are almost sensitive to noises. To overcome this problem, a new soft kernel convex hull support vector machine called SCH-SVM is proposed based on the soft kernel convex hull and pinball loss function. SCH-SVM extracts the soft convex hull vectors in the kernel space, which can represent geometric profile of data in the kernel space. Then SCH-SVM represents the original samples which projected as the soft convex hull vectors for the training samples, and finds the maximum quantile distance between soft kernel convex hulls belonging to two classes by using pinball loss function. Theoretical analysis and numerical experiments show that SCH-SVM has distinctive ability of training time, noise resistibility, and the number of support vectors.

Key words: large scale datasets, noise, soft kernel convex hull, pinball loss function, classification

中图分类号:

TP391.4

顾晓清, 倪彤光, 姜志彬, 等. 面向大规模噪声数据的软性核凸包支持向量机[J]. 电子学报, 2018, 46(2): 347-357.

GU Xiao-qing, NI Tong-guang, JIANG Zhi-bin, et al. Soft Kernel Convex Hull Support Vector Machine for Large Scale Noisy Datasets[J]. Acta Electronica Sinica, 2018, 46(2): 347-357.

参考文献

[1] Bo L F,Wang L L,Jiao C.Training hard-margin support vector machines using greedy stagewise algorithm[J].IEEE Trans.Neural Networks,2008,19(8):1446-1455.
[2] Fine S,Scheinberg K.Efficient SVM training using low-rank kernel representations[J].The Journal of Machine Learning Research,2001,2(3):243-264.
[3] Ni T G,Chung F L,Wang S T.Support vector machine with manifold regularization and partially labeling privacy protection[J].Information Sciences,2015,294(10):390-407.
[4] Tsang I W,Kwok J T,Cheung P M.Core vector machines:fast SVM training on very large data sets[J].The Journal of Machine Learning Research,2005,6(12):363-392.
[5] 胡文军,王士同.隐私保护的SVM快速分类方法[J].电子学报,2012,40(2):280-286. Hu Wenjun,Wang Shitong.Fast classification approach support vector machine with privacy preservation[J].Acta Electronica Sinica,2012,40(2):280-286.(in Chinese)
[6] Wang S T,Wang J,Chung F L.Kernel density estimation,kernel methods,and fast learning in large data sets[J].IEEE Trans.Cybernetics,2014,44(1):1-20.
[7] Nandan M,Khargonekar P P,Talathi S S.Fast SVM training using approximate extreme points[J].Journal of Machine Learning Research,2014,15:59-98.
[8] Kim W,Stankovc M S,Johansson K H,et al.A distributed support vector machine learning over wireless sensor networks[J].IEEE Trans Cybernetics,2015,45(11):2599-2611.
[9] Batuwita R,Palade V.FSVM-CIL:fuzzy support vector machines for class imbalance learning[J].IEEE Trans Fuzzy Systems,2010,18(3):558-571.
[10] Wang Y,Wang S,Lai K.A new fuzzy support vector machine to evaluate credit risk[J].IEEE Trans Fuzzy System,2005,13(6):820-831.
[11] An W J,Liang M G.Fuzzy support vector machine based on within-class scatter for classification problems with outliers or noises[J].Neurocomputing,2013,110(6):101-110.
[12] Huang X L,Shi L,Pelckmansb K,Suykens J A K.Asymmetric ν -tube support vector regression[J].Computational Statistics and Data Analysis,2014,77:371-382.
[13] Huang X L,Shi L,Suykens J A K.Support vector machine classifier with pinball loss[J].IEEE Trans Pattern Analysis and Machine Intelligence,2014,36(5):984-997.
[14] He X Y,Mourot G,Maquin D,et al.Multi-task learning with one-class SVM[J].Neurocomputing,2014,133:416-426.
[15] Wang D,Qiao H,Zhang B,et al.Online support vector machine based on convex hull vertices selection[J].IEEE Trans.Neural Networks and Learning Systems,2013,24(4):593-609.
[16] Dattorro J.Convex Optimization and Euclidean Distance Geometry[M].Mεβoo Publishing USA,2015.
[17] Blum M,Floyd R W,Pratt V R,et al.Time bounds for selection[J].Journal of Computer and System Sciences,1973,7(4):448-461.
[18] Xua J,Jiang Y X,Zeng C Q,et al.Node anomaly detection for homogeneous distributed environments[J].Expert Systems with Applications,2015,42(20):7012-7025.
[19] Tax D M J,Duin R P W.Support vector data description[J].Machine Learning,2004,54(1):45-66.
[20] Takahashi N,Nishi T.Rigorous proof of termination of SMO algorithm for support vector machines[J].IEEE Trans Neural Network,2005,16(3):774-776.
[21] Tsang I,Kwok A,Kwok J.Simpler core vector machines with enclosing balls[A].International conference on Machine learning[C].Corvallis,USA,2007.911-918.
[22] Bache K,Lichman M.UCI database[DB/OL].http://www.ics.uci.edu/%20mlearn/MLRepository.html.
[23] Luukka P,Lampinen J.Differential evolution classifier in noisy settings and with interacting variables[J].Applied Soft Computing,2011,11:891-899.
[24] Chang C,Lin C.LIBSVM:A library for support vector machines[J].ACM Trans Intelligence and System Technology,2011,2(3):1-27.

面向大规模噪声数据的软性核凸包支持向量机

Soft Kernel Convex Hull Support Vector Machine for Large Scale Noisy Datasets

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	郭庆, 陈雨庭, 段宗明, 吴先良. 基于65 nm CMOS工艺的小型化高增益低噪声放大器设计[J]. 电子学报, 2023, 51(3): 593-600.
[2]	饶伟, 贾凤勤, 李旦. 有色噪声下不相关和相干混合信号的DOA估计[J]. 电子学报, 2023, 51(3): 622-631.
[3]	吕杭, 蒋明峰, 李杨, 张鞠成, 王志康. 基于混合时频域特征的卷积神经网络心律失常分类方法的研究[J]. 电子学报, 2023, 51(3): 701-711.
[4]	王子为, 鲁继文, 周杰. 基于自适应梯度优化的二值神经网络[J]. 电子学报, 2023, 51(2): 257-266.
[5]	许新征, 李杉. 基于特征膨胀卷积模块的轻量化技术研究[J]. 电子学报, 2023, 51(2): 355-364.
[6]	王娜, 王俊平, 朱俊辉. 基于矩阵偏序关系的形态学算子[J]. 电子学报, 2023, 51(1): 213-221.
[7]	董健, 栾英宏, 赵永涛, 孙彦龙, 冯剑锋, 姜丽菲, 赵锋, 徐红新. 一维综合孔径辐射计相位定标研究[J]. 电子学报, 2022, 50(9): 2127-2133.
[8]	李雪莹, 王田路, 梁鹏, 王翀. 基于系统模型的用户评论中非功能需求的自动分类[J]. 电子学报, 2022, 50(9): 2079-2089.
[9]	徐叶, 张培勇, 李豪, 黄开天. 一种具有自适应优化电源抑制比的低静态电流无片外电容LDO[J]. 电子学报, 2022, 50(7): 1674-1683.
[10]	鲍加迪, 李云杰, 朱梦韬, 张蔚. 非理想观测下的多功能雷达工作状态在线切换点检测方法[J]. 电子学报, 2022, 50(6): 1291-1300.
[11]	吴懿霖, 曾珍, 章令杰, 张旨遥, 刘永. 基于双环结构的谐波锁模光电振荡器仿真模型[J]. 电子学报, 2022, 50(4): 789-795.
[12]	陆克中, 陈超凡, 蔡桓, 吴定明. 面向概念漂移和类不平衡数据流的在线分类算法[J]. 电子学报, 2022, 50(3): 585-597.
[13]	王雪松, 张翰林, 程玉虎. 基于自编码器和超图的半监督宽度学习系统[J]. 电子学报, 2022, 50(3): 533-539.
[14]	赵旭东, 陶然, 李伟, 张蒙蒙. 高光谱图像分数域信息提取理论与方法进展[J]. 电子学报, 2022, 50(12): 2874-2883.
[15]	孙晨峰, 吕卫民, 戴洪德, 张浩晨. 一种基于TimeGAN和OCSVM的多元退化设备小子样数据增广方法[J]. 电子学报, 2022, 50(11): 2678-2687.