This repository implements the CPNL method, proposed in our following paper:
Guoqiang Wu, Yingjie Tian, and Dalian Liu. "Cost-sensitive multi-label learning with positive and negative label pairwise correlations." Neural Networks 108 (2018): 411-423.

run the demo.m in MATLAB, CPNL will run as its default parameters on sample datasets.

If you have any problem with the code, please do not hesitate to contact Guoqiang Wu ([email protected]).

If you find this code is useful, please cite our paper:

@article{wu2018cost,
  title={Cost-sensitive multi-label learning with positive and negative label pairwise correlations},
  author={Wu, Guoqiang and Tian, Yingjie and Liu, Dalian},
  journal={Neural Networks},
  volume={108},
  pages={411--423},
  year={2018},
  publisher={Elsevier}
}