Empirical analysis of support vector machine ensemble classifiers |
| |
| Authors: | Shi-jin Wang Avin Mathew Yan Chen Li-feng Xi Lin Ma Jay Lee |
| |
| Affiliation: | 1. Department of Industrial Engineering and Management, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China;2. Cooperative Research Centre for Integrated Engineering Asset Management (CIEAM), Queensland University of Technology, Brisbane, Australia;3. NSF Center for Intelligent Maintenance Systems (IMS), University of Cincinnati, Cincinnati, USA;1. Department of Orthopedic Surgery, Rush University Medical Center, Chicago, Illinois, U.S.A.;2. Department of Orthopedic Surgery, University Hospitals Case Medical Center, Cleveland, Ohio, U.S.A.;1. Université de Tunis El Manar, Institut Pasteur de Tunis, Laboratory of Bioinformatics, Biomathematics & Biostatistics (BIMS), Tunis, Tunisia;2. Aix-Marseille Université, LISA, EA4672, Équipe METICA, 13397 Marseille cedex 20, France;3. Biochemistry Laboratory, LR12ES05 ‘Nutrition-Functional Food & Vascular Health’, USCR Mass Spectrometry, Faculty of Medicine, University of Monastir, Tunisia;4. Institut de l’Olivier, Unité Technologie et Qualité, BP 1087, 3018 Sfax, Tunisia;2. Shahrood University of Technology, Shahrood, Iran;1. Harbin Institute of Technology, Harbin, China;2. Bohai University, Jinzhou, China;1. Department of Computer Engineering, Chulalongkorn University, Pathumwan, Bangkok 10330, Thailand;2. School of Informatics, Walailak University, Thasala District, Nakhon Si Thammarat 80161, Thailand;1. Department of Computer Science and Artificial Intelligence, University of the Basque Country UPV/EHU, Donostia-San Sebastian 20018, Spain;2. Applied Mathematics Department, University of the Basque Country UPV/EHU, Donostia-San Sebastian 20018, Spain |
| |
| Abstract: | Ensemble classification – combining the results of a set of base learners – has received much attention in the machine learning community and has demonstrated promising capabilities in improving classification accuracy. Compared with neural network or decision tree ensembles, there is no comprehensive empirical research in support vector machine (SVM) ensembles. To fill this void, this paper analyses and compares SVM ensembles with four different ensemble constructing techniques, namely bagging, AdaBoost, Arc-X4 and a modified AdaBoost. Twenty real-world data sets from the UCI repository are used as benchmarks to evaluate and compare the performance of these SVM ensemble classifiers by their classification accuracy. Different kernel functions and different numbers of base SVM learners are tested in the ensembles. The experimental results show that although SVM ensembles are not always better than a single SVM, the SVM bagged ensemble performs as well or better than other methods with a relatively higher generality, particularly SVMs with a polynomial kernel function. Finally, an industrial case study of gear defect detection is conducted to validate the empirical analysis results. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
