Login
Register
Reset Password
Publish & Distribute
Publishing Solutions
Distribution Solutions
Subjects
Publications
Journals
Books
Proceedings
Publishers
Blog
Contact
Search
EUR
USD
GBP
English
English
Deutsch
Polski
Español
Français
Italiano
Cart
Home
Journals
Applied Mathematics and Nonlinear Sciences
Volume 8 (2023): Issue 1 (January 2023)
Open Access
AtanK-A New SVM Kernel for Classification
Meng Tian
Meng Tian
and
Hongkui Li
Hongkui Li
| Dec 31, 2021
Applied Mathematics and Nonlinear Sciences
Volume 8 (2023): Issue 1 (January 2023)
About this article
Previous Article
Next Article
Abstract
Figures & Tables
References
Authors
Articles in this Issue
Preview
PDF
Cite
Share
Published Online:
Dec 31, 2021
Page range:
135 - 146
Received:
Sep 04, 2020
Accepted:
Sep 27, 2021
DOI:
https://doi.org/10.2478/amns.2021.1.00092
Keywords
Arctangent kernel
,
Classification
,
Generalization
,
Kernel function
,
Support vector machine
© 2023 Meng Tian et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.
Fig. 1
Trends near the zero for AtanK kernel with different σ values.
Fig. 2
The trends near and far from the zero for Gauss, ERBF, UKF and AtanK kernels with different σ values.
Fig. 3
Two-dimensional Checkerboard dataset and classification boundaries of 5 sets of kernels on Checkerboard dataset.
Fig. 4
Boundary between domains of positive and negative samples corresponding to Cosexp dataset and the classification boundaries of 5 sets of kernels on Cosexp dataset.
Fig. 5
Result of Friedman’s test of the testing accuracy.