Institutional Repository
| a fuzzy support vector machine algorithm with dual membership based on hypersphere | |
| Ding Shifei; Gu Yaxiang | |
| 2011 | |
| Source | Journal of Computational Information Systems
 |
| ISSN | 1553-9105 |
| Volume | 7Issue:6Pages:2028-2034 |
| English Abstract | In traditional fuzzy support vector machine(FSVM), membership function is established in global scope will reduce the membership of support vectors, and the FSVM based dismissing margin increases the training speed, but will remove some support vector artificially. So, a new algorithm of Fuzzy Support Vector Machine with Dual Membership based on Hypersphere (HDM-FSVM) is proposed. In this algorithm, the two classes of hyperspheres are divided into two parts respectively. Then, according to most support vectors are in the hemispheres which close together, we use the membership function that can enhance the membership of support vector, and because of there are a few of support vectors in other hemispheres, we must ensure the high membership of support vectors and reduce the membership of non-support vector. In order to removal noise and outliers, we introduce a radius controlling factor to control size of hyperspheres, the samples that outside of hyperspheres are considered as noise and outliers. Experimental results show that HDM-FSVM can enhance the classification accuracy rate of the sample sets that contain noise and outliers. Copyright © 2011 Binary Information Press.; In traditional fuzzy support vector machine(FSVM), membership function is established in global scope will reduce the membership of support vectors, and the FSVM based dismissing margin increases the training speed, but will remove some support vector artificially. So, a new algorithm of Fuzzy Support Vector Machine with Dual Membership based on Hypersphere (HDM-FSVM) is proposed. In this algorithm, the two classes of hyperspheres are divided into two parts respectively. Then, according to most support vectors are in the hemispheres which close together, we use the membership function that can enhance the membership of support vector, and because of there are a few of support vectors in other hemispheres, we must ensure the high membership of support vectors and reduce the membership of non-support vector. In order to removal noise and outliers, we introduce a radius controlling factor to control size of hyperspheres, the samples that outside of hyperspheres are considered as noise and outliers. Experimental results show that HDM-FSVM can enhance the classification accuracy rate of the sample sets that contain noise and outliers. Copyright © 2011 Binary Information Press. |
| Indexed Type | EI |
| Keyword | Algorithms Statistics Support Vector Machines Vectors |
| Department | (1) School of Computer Science and Technology China University of Mining and Technology Xuzhou 221116 China; (2) Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Science Beijing 100080 China; (3) Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia Beijing University of Posts and Telecommunications Beijing 100876 China |
| Language | 英语 |
| Content Type | 期刊论文 |
| URI | http://ir.iscas.ac.cn/handle/311060/16171 |
| Collection | 中国科学院软件研究所 |
| Recommended Citation GB/T 7714 | Ding Shifei,Gu Yaxiang. a fuzzy support vector machine algorithm with dual membership based on hypersphere[J]. Journal of Computational Information Systems,2011,7(6):2028-2034. |
| APA | Ding Shifei,&Gu Yaxiang.(2011).a fuzzy support vector machine algorithm with dual membership based on hypersphere.Journal of Computational Information Systems,7(6),2028-2034. |
| MLA | Ding Shifei,et al."a fuzzy support vector machine algorithm with dual membership based on hypersphere".Journal of Computational Information Systems 7.6(2011):2028-2034. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment