Description:
WellSVM is a package for scalable and convex weakly labeled SVMs. The package includes the MATLAB code of the algorithm WellSVM.
Reference:
Yu-Feng Li, Ivor Tsang, James Kwok and Zhi-Hua Zhou.
Scalable and convex weakly labeled SVMs. Journal of Machine Learning Research, 14:2151-2188, 2013.
ATTN:
This package is free for academic usage. You can run it at your own risk. For other purposes, please contact
Prof. Zhi-Hua Zhou (zhouzh@nju.edu.cn).
Requirement:
The package was developed with MATLAB(v2008a) and C++, hence Matlab and C++ need to be installed before running this code. A demo function and some data sets are included to illustrate the use of this package.
ATTN2:
This package was developed by
Mr.Yu-Feng Li (liyf@lamda.nju.edu.cn). For any problem concerning the code, please feel free to contact Mr. Li.
Download:
code & data(2.7 MB), for semi-supervised learning, regular scale data sets
code & data(40 MB), for semi-supervised learning, large scale data sets
code & data(814 KB), for clustering, regular scale data sets
code(141 KB) &
data(456 MB), for clustering, large scale data sets
code for multiple instance learning