A new multi-class classification method based on minimum enclosing balls

详细信息    查看全文

• 作者：QingJun Song ; XingMing Xiao ; HaiYan Jiang…
• 关键词：Support vector machine ; Multi ; class classification ; Minimum enclosing balls ; Gaussian kernel ; Width factor
• 刊名：Journal of Mechanical Science and Technology
• 出版年：2015
• 出版时间：August 2015
• 年：2015
• 卷：29
• 期：8
• 页码：3467-3473
• 全文大小：342 KB
• 参考文献：[1]J. C. B. Christopher, A tutorial on support vector machines for pattern recognition, Data Mining and Knowledge Discovery, 2 (1998) 121-67.View Article
  [2]J. L. Balcázar, Y. Dai, J. Tanaka and O. Watanabe, Provably fast training algorithms for support vector machines, Theory Comput Syst, 42 (2008) 568-95.View Article
  [3]N. Bassam et al., Utilizing wavelet transform and support vector machine for detection of the paradoxical splitting in the second heart sound, Med. Biol. Eng. Comput., 48 (2010) 177-84.View Article
  [4]N. Collier et al., Learning to recognize phenotype candidates in the auto-immune literature using SVM re-ranking, Plos One, 8 (10) (2013) 1-5.View Article
  [5]R. C. Concei??o et al., Evaluation of features and classifiers for classification of early-stage breast cancer, J. of Electromagn. Waves and App, 25 (2011) 1-4.View Article
  [6]L. L. Diao, C. H. Z. H. Yang and H. Wang, Training SVM email classifiers using very large imbalanced dataset, J. Exp. Theor. Artif. In., 24 (2) (2012) 193-10.View Article
  [7]J. Mashford et al., Prediction of sewer condition grade using support vector machines, J. Comput. Civil Eng., 25 (4) (2011) 283-90.View Article
  [8]B. Zaman, M. Mckee and C. M. U. Neale, Fusion of remotely sensed data for soil moisture estimation using relevance vector and support vector machines, Int. J. Remote Sens., 33 (20) (2012) 6516-552.View Article
  [9]A. K. Verma and T. N. Singh, Comparative study of cognitive systems for ground vibration measurements, Neural Compu.t & Applic., 22 (2013) S341–S350.View Article
  [10]W. F. Cheng and Ch. B. Guang, Wavelet correlation feature scale entropy and fuzzy support vector machine approach for aeroengine whole-body vibration fault diagnosis, Shock and Vibration, 20 (2013) 341-49.View Article
  [11]W. T. Mao, D. K. Hu and G. R. Yan, A new SVM regression approach for mechanical load identification, Int. J. Appl. Electrom., 33 (2010) 1001-008.
  [12]O. Kadri, L. H. Mouss and M. D. Mouss, Fault diagnosis of rotary kiln using SVM and binary ACO, Journal of Mechanical Science and Technology, 26 (2) (2012) 601-08.View Article
  [13]T. Kamali, R. Boostani and H. Parsaei, A multi-classifier approach to MUAP classification for diagnosis of neuromuscular disorders, IEEE T. Neur Sys. Reh., 22 (1) (2014) 191-00.View Article
  [14]J. Amores, N. Sebe and P. Radeva, Context-based objectclass recognition and retrieval by generalized correlograms, IEEE T. Pattern AnalL, 29 (2007) 1818-833.View Article
  [15]M. Jian and Ch. K. Jung, Class-discriminative kernel sparse representation-based classification using multi-objective optimization, IEEE T. Signal Proces, 61 (18) (2013) 4416-427.View Article
  [16]L. P. Jing and K. N. Michael, Sparse label-indicator optimization methods for image classification, IEEE T. Image Processing, 23 (3) (2014) 1002-014.View Article
  [17]M. Mohammadi and G. B. Gharehpetian, Application of multi-class support vector machines for power system online static security assessment using DT-based feature and data selection algorithms, J. Intell. Fuzzy Syst., 20 (2009) 133-46.
  [18]X. L. Tang et al., Multi-fault pattern recognition based on support vector machine trained by chaos particle swarm optimization, Knowl.-based Syst., 23 (2010) 486-90.View Article
  [19]P. S. Kang and S. Z. Cho, Support vector class description (SVCD): Classification in kernel space, Intelligent Data Analysis, 16 (2012) 351-64.
  [20]T. Nakagawa, Y. Iwahori and M. K. Bhuyan, Defect classification of electronic board using multiple classifiers and grid search of SVM parameters, Computer and Information Scence, 493 (2013) 115-27.View Article
  [21]E. Welzl. Smallest enclosing disks (balls and ellipsoids), New Results and New Trends in Computer Science, 555 (1991) 359-70.View Article
  [22]I. Tsang, J. Kwok and P. M. Cheung, Core vector machines: fast SVM training on very large data sets, J. Mach. Learn Res., 6 (2005) 363-92.
  [23]P. Y. Hao, J. H. Chiang and Y. H. Lin, A new maximalmargin spherical-structured multi-class support vector machine, Appl. Intell., 30 (2009) 98-11.View Article
  [24]F. F. Li, B. Yao and P. Kumar, Group enclosing queries, IEEE T Knowl. Data EN, 23 (100) (2011) 1526-540.View Article
  [25]SH. F. Yuan and M. Li, Fault diagnosis using binary tree and sphere-structured support vector machines, Journal of Mechanical Science and Technology, 26 (5) (2012) 1431-438.View Article
  [26]T. Le et al., Proximity multi-sphere support vector clustering, Neural Comput. & Applic., 22 (2013) 1309-319.View Article
  [27]J. G. Wang, P. Neskovic and L. N. Cooper, Bayes classification based on minimum bounding spheres, Neurocomputing, 70 (2007) 801-08.View Article
  [28]E. Frandi et al., Training support vector machines using Frank-Wolfe optimization methods, Int. J. Pattern. Recogn., 27 (3) (2013) 1360003.1-40.
  [29]F. L. Chung, Zh. H. Deng and Sh. T. Wang, From minimum encl
• 作者单位：QingJun Song (1) (2)
  XingMing Xiao (1)
  HaiYan Jiang (2)
  XieGuang Zhao (2)
  
  1. School of Mechanical and Electrical Engineering, China University of Mining & Technology, Xuzhou, Jiangsu, 221008, China
  2. School of Tai-an, Shandong University of Science & Technology, Tai-an, Shandong, 271019, China
  
• 刊物类别：Engineering
• 刊物主题：Mechanical Engineering
  Structural Mechanics
  Control Engineering
  Industrial and Production Engineering
  
• 出版者：The Korean Society of Mechanical Engineers
• ISSN：1976-3824

文摘

With respect to classification problems, the Minimum enclosing ball (MEB) method was recently studied by some scholars as a new support vector machine. As a nascent technology, however, MEB reports poor adaptability for different types of samples, especially multi-class samples. In this paper, we propose a new multi-class classification method based on MEB. This method is derived from each class sample center and radius with the Gaussian kernel width factor parameter σ, which is labelled as σ-MEB. σ is a variable parameter according to the different sample characteristics. When this parameter is considered, the multi-class classifier is easy to adapt and is robust in diverse datasets. The quadratic programming problem was transformed into its dual form with Lagrange multipliers using this method. Finally, we applied sequential minimal optimization method and Karush—Kuhn—Tucker conditions to accelerate the training process. Numerical experiment results indicate that for given different types of samples, the proposed method is more accurate than the methods with which it is compared. Moreover, the proposed method reports values in the upper quantile with respect to adaptation performance.