A Unified View of Multi-Label Performance Measures

Xi-Zhu Wu, Zhi-Hua Zhou
Proceedings of the 34th International Conference on Machine Learning, PMLR 70:3780-3788, 2017.

Abstract

Multi-label classification deals with the problem where each instance is associated with multiple class labels. Because evaluation in multi-label classification is more complicated than single-label setting, a number of performance measures have been proposed. It is noticed that an algorithm usually performs differently on different measures. Therefore, it is important to understand which algorithms perform well on which measure(s) and why. In this paper, we propose a unified margin view to revisit eleven performance measures in multi-label classification. In particular, we define label-wise margin and instance-wise margin, and prove that through maximizing these margins, different corresponding performance measures are to be optimized. Based on the defined margins, a max-margin approach called LIMO is designed and empirical results validate our theoretical findings.

Cite this Paper


BibTeX
@InProceedings{pmlr-v70-wu17a, title = {A Unified View of Multi-Label Performance Measures}, author = {Xi-Zhu Wu and Zhi-Hua Zhou}, booktitle = {Proceedings of the 34th International Conference on Machine Learning}, pages = {3780--3788}, year = {2017}, editor = {Precup, Doina and Teh, Yee Whye}, volume = {70}, series = {Proceedings of Machine Learning Research}, month = {06--11 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v70/wu17a/wu17a.pdf}, url = {https://proceedings.mlr.press/v70/wu17a.html}, abstract = {Multi-label classification deals with the problem where each instance is associated with multiple class labels. Because evaluation in multi-label classification is more complicated than single-label setting, a number of performance measures have been proposed. It is noticed that an algorithm usually performs differently on different measures. Therefore, it is important to understand which algorithms perform well on which measure(s) and why. In this paper, we propose a unified margin view to revisit eleven performance measures in multi-label classification. In particular, we define label-wise margin and instance-wise margin, and prove that through maximizing these margins, different corresponding performance measures are to be optimized. Based on the defined margins, a max-margin approach called LIMO is designed and empirical results validate our theoretical findings.} }
Endnote
%0 Conference Paper %T A Unified View of Multi-Label Performance Measures %A Xi-Zhu Wu %A Zhi-Hua Zhou %B Proceedings of the 34th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2017 %E Doina Precup %E Yee Whye Teh %F pmlr-v70-wu17a %I PMLR %P 3780--3788 %U https://proceedings.mlr.press/v70/wu17a.html %V 70 %X Multi-label classification deals with the problem where each instance is associated with multiple class labels. Because evaluation in multi-label classification is more complicated than single-label setting, a number of performance measures have been proposed. It is noticed that an algorithm usually performs differently on different measures. Therefore, it is important to understand which algorithms perform well on which measure(s) and why. In this paper, we propose a unified margin view to revisit eleven performance measures in multi-label classification. In particular, we define label-wise margin and instance-wise margin, and prove that through maximizing these margins, different corresponding performance measures are to be optimized. Based on the defined margins, a max-margin approach called LIMO is designed and empirical results validate our theoretical findings.
APA
Wu, X. & Zhou, Z.. (2017). A Unified View of Multi-Label Performance Measures. Proceedings of the 34th International Conference on Machine Learning, in Proceedings of Machine Learning Research 70:3780-3788 Available from https://proceedings.mlr.press/v70/wu17a.html.

Related Material