Email this article A New Fuzzy SVM based on the Posterior Probability Weighting Membership
Abstract
To solve the sensitivity to the noises and outliers in support vector machine (SVM), the characterizations of fuzzy support vector machine (FSVM) are analyzed. But the determination of fuzzy membership is a difficulty. By the inspiration of bayesian decision theory and combining with sample density to give weight for each sample, new fuzzy membership function is proposed. Each sample points is given the tightness arranged forecasts by this method and the generalization ability of FSVM is improved. Numerical experiments show that, compared with the traditional SVM and FSVM, the improved algorithm performs, more effectively and accurately, has better classification result.
Keywords
References
[1] Gu Ya-xiang, Ding Shi-fei, “Advances of Support Vector Machine”, Computer Science, vol. 38, no. 2, pp. 14-17, 2011.
[2] C. F. Lin, S. D. Wang, “Fuzzy Support Vector Machines”, IEEE Transactions on Neural Networks, vol.13, no. 2, pp. 464-471, 2002.
[3] H. P. Huang, Y. H. Liu, “Fuzzy Support Vector Machines for Pattern Recognition and Data Mining”, International Journal of Fuzzy Systems, vol.4, no. 3, pp. 826-835, 2002.
[4] Hao Tang, Yuhe Liao, Feng Sun, Xie Hang, “The FSVM Algorithm with Fuzzy Membership”, Journal of Xi’an Jiaotong University, vol.43, no.7, pp. 40-43, 2009.
[5] Ying Zhang, Hongye Sun, Jian Zhu, “Fuzzy Support Vector Regression Based on Data Domain Description”, Journal of Information and Control, vol. 34, no. 1, pp. 1-6, 2005.
PMid:19288593
[6] Hsu Che-Chang, Han Ming-Feng, Chang Shih-Hsing, Chung Hung-Yuan, “Fuzzy support vector machines with the uncertainty of parameter C”, Expert Systems with Applications, vol. 36, Issue 3, pp. 6654-6658, 2009.
[7] Li Miao-miao, Xiang Feng-hong, Liu Xin-wang, “A Novel Membership Function for Fuzzy Support Vector Machines”, Computer Engineering & Science, vol. 31, no. 9, pp. 92-94, 2009.
[8] Li Lei, Zhou Meng-meng, Lu Yan-ling, “Fuzzy Support Vector Machine Based on Density with Dual Membership”, Computer Technology and Development, vol.19, no. 12, pp. 44-46, 2009.
[9] Gaowei Wu, Qin Tao, Yu Wang, “Support Vector Machines Based on Posterior Probability”, Journal of Computer Research and Development, vol. 42, no. 2, pp. 196-202, 2005.
[10] Zhaoqi Bian, Xuegong Zhang, Pattern Recognition (Second Edition), Beijing: Tsinghua University Press, 2000.
[11] Mengcehng Yao, Probability and Statistics Tutorial, Beijing: Tsinghua University Press, 2007.
[12] Yong Zhang, Zhong-Xian Chi, “A Fuzzy support vector classifier based on Bayesian optimization”, Fuzzy Optim Decis Making, vol. 7, pp. 75–86, 2008.
[13] Huaji Shi, Shuyong Zhou, Xingyi Li, Hui Tang, Qiulin Ding, “To Detect Outlier Based on the Average Density”, Journal of University of Electronic Science and Technology of China, vol. 36, no. 6, pp. 1286-1288, 2007.
[14] Hao Tang, Liang-sheng Qu, “Fuzzy support vector machine with a new fuzzy membership function for pattern classification”, Proceedings of the Seventh International Conference on Machine Learning and Cybernetics, pp. 768-773, 2008.
[15] Nemmour Hassiba, Chibani Youcef, “Fuzzy integral to speed up support vector machines training for pattern classification”, International Journal of Knowledge Based Intelligent Engineering Systems, vol. 14, Issue 3, pp. 127-138, 2010.
Full Text: PDF