Robust label compression for multi-label classification

作者：

Highlights：

• This paper deals with label compression of multi-label classification.

• It is the first paper considering outliers in label compression.

• Outliers in the feature space are taken into account.

• Irregular label correlations can also be thought as outliers.

• This paper tackles this problem by using l2,1-norm.

摘要

•This paper deals with label compression of multi-label classification.•It is the first paper considering outliers in label compression.•Outliers in the feature space are taken into account.•Irregular label correlations can also be thought as outliers.•This paper tackles this problem by using l2,1-norm.

论文关键词：Multi-label classification,Label compression,Encoding loss,Dependence loss,Outliers,l2,1-norm

论文评审过程：Received 19 October 2015, Revised 24 May 2016, Accepted 25 May 2016, Available online 29 May 2016, Version of Record 9 July 2016.

论文官网地址：https://doi.org/10.1016/j.knosys.2016.05.051