SVM Classification Model Parameters Optimized by Improved Genetic Algorithm

Qing Hua Mao; Hong Wei Ma; Xu Hui Zhang

doi:10.4028/www.scientific.net/AMR.889-890.617

Paper Titles

Structural Topology Optimization Using Turing Model
p.595

New Algorithms for Solving Minimum Path Sets of Fault Tree Analysis and its Applications
p.600

Multi-Objective Model and Heuristic Algorithm for Billet Stacking Problem in Steel Production
p.606

Using Mathematical Programming Optimization Work Exchange Networks
p.612

SVM Classification Model Parameters Optimized by Improved Genetic Algorithm
p.617

A New Heuristic Topology Optimization Method Based on Bone Remodeling Model
p.622

The Traveling Wave Solutions of Nonlinear Beam Equations by Tanh-Function Method
p.628

Research on Impulsive Lurie Systems with Time-Delay
p.633

Loader Fatigue Correlation Analysis on the Lab Conditions and Theoretical Calculation
p.637

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 889-890SVM Classification Model Parameters Optimized by...

SVM Classification Model Parameters Optimized by Improved Genetic Algorithm

Abstract:

SVM classification model has been widely applied to mechanical equipment fault diagnosis and material defects classification. It is difficult to choose the optimal value of penalty factor C and kernel function parameter for SVM model. Therefore, an improved genetic algorithm to optimize SVM parameters is put forward, which improves crossover and mutation operators and enhances convergence properties by using the best individual retention strategy. UCI data set is used to verify the algorithm. The testing results show that the algorithm can quickly and effectively select optimal SVM parameters and improve SVM classification accuracy.

You might also be interested in these eBooks

Engineering Solutions for Manufacturing Processes IV

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 889-890)

Pages:

617-621

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.889-890.617

Citation:

Cite this paper

Online since:

February 2014

Authors:

Qing Hua Mao, Hong Wei Ma, Xu Hui Zhang

Keywords:

Classification, Improved Genetic Algorithm (IGA), Optimization, SVM

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Chen Guo. Optimizing the Parameters of Support Vector Machine Classifier Model Bas-ed on Genetic Algorithm[J]. Mechanical Science and Technology, 2007, 26(3): 347-350(In Chinese).

Google Scholar

[2] Wan Yuan, Tong Hengqing, Zhu Yingying. Parameters Optimization of Multi-Kernel Su-pport Vector Machine Based on Genetic Algorithm[J]. Journal of Wuhan University 2012, 58(3): 255-259(In Chinese).

Google Scholar

[3] Wang Jiao, Liu Haiyan. Support Vector Machine and Genetic Algorithm based Tool Fai-lure Diagnosis[J]. Modular Machine Tool ＆ Automatic Manufacturing Technique, 2013, (1): 74-76(In Chinese).

Google Scholar

[4] Chen Shijie, Lian Ke, Wang Houjun. Method for Analog Circuit Fault Diagnosis Based on GA Optimized SVM[J]. Journal of University of Electronic Science and Technology of China, 2009, 38(4): 553-558(In Chinese).

Google Scholar

[5] Daohong Qiu, Shucai Li, Lewen Zhang, Yiguo Xue. Application of GASVM in Classif-icationof Surrounding Rock Based on Model Reliability Examination [J]. Mining Science and Technology (China), 2010, 20(3): 428-433(In Chinese).

DOI: 10.1016/s1674-5264(09)60221-0

Google Scholar

[6] Han Han, Houjun Wang and etc. Transformer Fault Dignosis Based on Feature Selectionand Parameter Optimization[J]. Energy Procedia, 2011, 12: 662-668.

DOI: 10.1016/j.egypro.2011.10.090

Google Scholar