This package is a MATLAB implementation of "Coordinate Descent Fuzzy Twin Support Vector Machine for Classification", Bin-Bin Gao, Jian-Jun Wang, Yao Wang and Chan-Yun Yang. You can train cdftsvm model on your own dataset for binary classification problems. This package is created by Bin-Bin Gao.
The CDFTSVM not only inherits the advantages of twin SVM but also leads to a fast and robust classification results. It is suitable for high dimension and noisy inputs.
In matlab, run demo.m.
If you find CDFTSVM helpful, please cite it as
@inproceedings{gao2015coordinate,
title={Coordinate Descent Fuzzy Twin Support Vector Machine for Classification},
author={Gao, Bin-Bin and Wang, Jian-Jun and Wang, Yao and Yang, Chan-Yun},
booktitle={2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA)},
pages={7--12},
year={2015},
organization={IEEE}
}
For any questions and comments, please send your email to csgaobb@gmail.com.
ATTN1: This packages are free for academic usage. You can run them at your own risk. For other purposes, please contact Bin-Bin Gao (csgaobb@gmail.com).
ATTN2: This packages were developed by Bin-Bin Gao. For any problem concerning the code, please feel free to contact Bin-Bin Gao (csgaobb@gmail.com).
version: Sep. 16, 2017