Hierarchical Multi-Label Classification Networks

Jonatas Wehrmann, Ricardo Cerri, Rodrigo Barros
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:5075-5084, 2018.

Abstract

One of the most challenging machine learning problems is a particular case of data classification in which classes are hierarchically structured and objects can be assigned to multiple paths of the class hierarchy at the same time. This task is known as hierarchical multi-label classification (HMC), with applications in text classification, image annotation, and in bioinformatics problems such as protein function prediction. In this paper, we propose novel neural network architectures for HMC called HMCN, capable of simultaneously optimizing local and global loss functions for discovering local hierarchical class-relationships and global information from the entire class hierarchy while penalizing hierarchical violations. We evaluate its performance in 21 datasets from four distinct domains, and we compare it against the current HMC state-of-the-art approaches. Results show that HMCN substantially outperforms all baselines with statistical significance, arising as the novel state-of-the-art for HMC.

Cite this Paper

BibTeX
@InProceedings{pmlr-v80-wehrmann18a, title = {Hierarchical Multi-Label Classification Networks}, author = {Wehrmann, Jonatas and Cerri, Ricardo and Barros, Rodrigo}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {5075--5084}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/wehrmann18a/wehrmann18a.pdf}, url = {https://proceedings.mlr.press/v80/wehrmann18a.html}, abstract = {One of the most challenging machine learning problems is a particular case of data classification in which classes are hierarchically structured and objects can be assigned to multiple paths of the class hierarchy at the same time. This task is known as hierarchical multi-label classification (HMC), with applications in text classification, image annotation, and in bioinformatics problems such as protein function prediction. In this paper, we propose novel neural network architectures for HMC called HMCN, capable of simultaneously optimizing local and global loss functions for discovering local hierarchical class-relationships and global information from the entire class hierarchy while penalizing hierarchical violations. We evaluate its performance in 21 datasets from four distinct domains, and we compare it against the current HMC state-of-the-art approaches. Results show that HMCN substantially outperforms all baselines with statistical significance, arising as the novel state-of-the-art for HMC.} }
Endnote
%0 Conference Paper %T Hierarchical Multi-Label Classification Networks %A Jonatas Wehrmann %A Ricardo Cerri %A Rodrigo Barros %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-wehrmann18a %I PMLR %P 5075--5084 %U https://proceedings.mlr.press/v80/wehrmann18a.html %V 80 %X One of the most challenging machine learning problems is a particular case of data classification in which classes are hierarchically structured and objects can be assigned to multiple paths of the class hierarchy at the same time. This task is known as hierarchical multi-label classification (HMC), with applications in text classification, image annotation, and in bioinformatics problems such as protein function prediction. In this paper, we propose novel neural network architectures for HMC called HMCN, capable of simultaneously optimizing local and global loss functions for discovering local hierarchical class-relationships and global information from the entire class hierarchy while penalizing hierarchical violations. We evaluate its performance in 21 datasets from four distinct domains, and we compare it against the current HMC state-of-the-art approaches. Results show that HMCN substantially outperforms all baselines with statistical significance, arising as the novel state-of-the-art for HMC.
APA
Wehrmann, J., Cerri, R. & Barros, R.. (2018). Hierarchical Multi-Label Classification Networks. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:5075-5084 Available from https://proceedings.mlr.press/v80/wehrmann18a.html.

Related Material