Optimization model and application of linear and nonlinear MBSVM based on pinball loss function
The multiple birth support vector machine (MBSVM) typically utilizes a hinge loss function, which is susceptible to noise sensitivity and instability during resampling. To overcome these issues, we introduce two models: linear and nonlinear MBSVM models, both utilizing a pinball loss function. The m...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2024-12-01
|
| Series: | Mathematical and Computer Modelling of Dynamical Systems |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/13873954.2024.2424854 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846126630480642048 |
|---|---|
| author | Linfeng Dai Longwei Chen Min Luo |
| author_facet | Linfeng Dai Longwei Chen Min Luo |
| author_sort | Linfeng Dai |
| collection | DOAJ |
| description | The multiple birth support vector machine (MBSVM) typically utilizes a hinge loss function, which is susceptible to noise sensitivity and instability during resampling. To overcome these issues, we introduce two models: linear and nonlinear MBSVM models, both utilizing a pinball loss function. The main goal of these models is to improve the classification performance and noise resilience of the Pin-MBSVM by optimizing the maximum quantile distance. Additionally, to decrease the computation time for these models, we provide an in-depth discussion of a fast algorithm based on the successive overrelaxation (SOR) iteration method. In experiments, we test the Pin-MBSVM algorithm using both UCI datasets and synthetic datasets, comparing its performance with OVO-TWSVM, OVA-TWSVM and MBSVM. The results show that our methods successfully reduce the noise sensitivity and resampling instability found in traditional MBSVM while preserving the model’s computational efficiency. Finally, the effectiveness of our model was validated using the Friedman test and Bonferroni–Dunn test. |
| format | Article |
| id | doaj-art-212693c400df4e4a980c318cc6b6f3d2 |
| institution | Kabale University |
| issn | 1387-3954 1744-5051 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Mathematical and Computer Modelling of Dynamical Systems |
| spelling | doaj-art-212693c400df4e4a980c318cc6b6f3d22024-12-12T14:07:39ZengTaylor & Francis GroupMathematical and Computer Modelling of Dynamical Systems1387-39541744-50512024-12-0130189892310.1080/13873954.2024.2424854Optimization model and application of linear and nonlinear MBSVM based on pinball loss functionLinfeng Dai0Longwei Chen1Min Luo2Department of Applied Mathematics, Yunnan University of Finance and Economics, KunMing, YunNan, People’s Republic of ChinaYunnan Key Laboratory of Service Computing, Yunnan University of Finance and Economics, Kunming, People’s Republic of ChinaDepartment of Applied Mathematics, Yunnan University of Finance and Economics, KunMing, YunNan, People’s Republic of ChinaThe multiple birth support vector machine (MBSVM) typically utilizes a hinge loss function, which is susceptible to noise sensitivity and instability during resampling. To overcome these issues, we introduce two models: linear and nonlinear MBSVM models, both utilizing a pinball loss function. The main goal of these models is to improve the classification performance and noise resilience of the Pin-MBSVM by optimizing the maximum quantile distance. Additionally, to decrease the computation time for these models, we provide an in-depth discussion of a fast algorithm based on the successive overrelaxation (SOR) iteration method. In experiments, we test the Pin-MBSVM algorithm using both UCI datasets and synthetic datasets, comparing its performance with OVO-TWSVM, OVA-TWSVM and MBSVM. The results show that our methods successfully reduce the noise sensitivity and resampling instability found in traditional MBSVM while preserving the model’s computational efficiency. Finally, the effectiveness of our model was validated using the Friedman test and Bonferroni–Dunn test.https://www.tandfonline.com/doi/10.1080/13873954.2024.2424854Multi-class classificationMBSVMpinball loss function |
| spellingShingle | Linfeng Dai Longwei Chen Min Luo Optimization model and application of linear and nonlinear MBSVM based on pinball loss function Mathematical and Computer Modelling of Dynamical Systems Multi-class classification MBSVM pinball loss function |
| title | Optimization model and application of linear and nonlinear MBSVM based on pinball loss function |
| title_full | Optimization model and application of linear and nonlinear MBSVM based on pinball loss function |
| title_fullStr | Optimization model and application of linear and nonlinear MBSVM based on pinball loss function |
| title_full_unstemmed | Optimization model and application of linear and nonlinear MBSVM based on pinball loss function |
| title_short | Optimization model and application of linear and nonlinear MBSVM based on pinball loss function |
| title_sort | optimization model and application of linear and nonlinear mbsvm based on pinball loss function |
| topic | Multi-class classification MBSVM pinball loss function |
| url | https://www.tandfonline.com/doi/10.1080/13873954.2024.2424854 |
| work_keys_str_mv | AT linfengdai optimizationmodelandapplicationoflinearandnonlinearmbsvmbasedonpinballlossfunction AT longweichen optimizationmodelandapplicationoflinearandnonlinearmbsvmbasedonpinballlossfunction AT minluo optimizationmodelandapplicationoflinearandnonlinearmbsvmbasedonpinballlossfunction |