Multi-label Classification with Error-correcting Codes

Chung-Sung Ferng, Hsuan-Tien Lin
Proceedings of the Asian Conference on Machine Learning, PMLR 20:281-295, 2011.

Abstract

We formulate a framework for applying error-correcting codes (ECC) on multi-label classification problems. The framework treats some base learners as noisy channels and uses ECC to correct the prediction errors made by the learners. An immediate use of the framework is a novel ECC-based explanation of the popular random k-label-sets (RAKEL) algorithm using a simple repetition ECC. Using the framework, we empirically compare a broad spectrum of ECC designs for multi-label classification. The results not only demonstrate that RAKEL can be improved by applying some stronger ECC, but also show that the traditional Binary Relevance approach can be enhanced by learning more parity-checking labels. In addition, our study on different ECC helps understand the trade-off between the strength of ECC and the hardness of the base learning tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v20-ferng11, title = {Multi-label Classification with Error-correcting Codes}, author = {Ferng, Chung-Sung and Lin, Hsuan-Tien}, booktitle = {Proceedings of the Asian Conference on Machine Learning}, pages = {281--295}, year = {2011}, editor = {Hsu, Chun-Nan and Lee, Wee Sun}, volume = {20}, series = {Proceedings of Machine Learning Research}, address = {South Garden Hotels and Resorts, Taoyuan, Taiwain}, month = {14--15 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v20/ferng11/ferng11.pdf}, url = {https://proceedings.mlr.press/v20/ferng11.html}, abstract = {We formulate a framework for applying error-correcting codes (ECC) on multi-label classification problems. The framework treats some base learners as noisy channels and uses ECC to correct the prediction errors made by the learners. An immediate use of the framework is a novel ECC-based explanation of the popular random k-label-sets (RAKEL) algorithm using a simple repetition ECC. Using the framework, we empirically compare a broad spectrum of ECC designs for multi-label classification. The results not only demonstrate that RAKEL can be improved by applying some stronger ECC, but also show that the traditional Binary Relevance approach can be enhanced by learning more parity-checking labels. In addition, our study on different ECC helps understand the trade-off between the strength of ECC and the hardness of the base learning tasks.} }
Endnote
%0 Conference Paper %T Multi-label Classification with Error-correcting Codes %A Chung-Sung Ferng %A Hsuan-Tien Lin %B Proceedings of the Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2011 %E Chun-Nan Hsu %E Wee Sun Lee %F pmlr-v20-ferng11 %I PMLR %P 281--295 %U https://proceedings.mlr.press/v20/ferng11.html %V 20 %X We formulate a framework for applying error-correcting codes (ECC) on multi-label classification problems. The framework treats some base learners as noisy channels and uses ECC to correct the prediction errors made by the learners. An immediate use of the framework is a novel ECC-based explanation of the popular random k-label-sets (RAKEL) algorithm using a simple repetition ECC. Using the framework, we empirically compare a broad spectrum of ECC designs for multi-label classification. The results not only demonstrate that RAKEL can be improved by applying some stronger ECC, but also show that the traditional Binary Relevance approach can be enhanced by learning more parity-checking labels. In addition, our study on different ECC helps understand the trade-off between the strength of ECC and the hardness of the base learning tasks.
RIS
TY - CPAPER TI - Multi-label Classification with Error-correcting Codes AU - Chung-Sung Ferng AU - Hsuan-Tien Lin BT - Proceedings of the Asian Conference on Machine Learning DA - 2011/11/17 ED - Chun-Nan Hsu ED - Wee Sun Lee ID - pmlr-v20-ferng11 PB - PMLR DP - Proceedings of Machine Learning Research VL - 20 SP - 281 EP - 295 L1 - http://proceedings.mlr.press/v20/ferng11/ferng11.pdf UR - https://proceedings.mlr.press/v20/ferng11.html AB - We formulate a framework for applying error-correcting codes (ECC) on multi-label classification problems. The framework treats some base learners as noisy channels and uses ECC to correct the prediction errors made by the learners. An immediate use of the framework is a novel ECC-based explanation of the popular random k-label-sets (RAKEL) algorithm using a simple repetition ECC. Using the framework, we empirically compare a broad spectrum of ECC designs for multi-label classification. The results not only demonstrate that RAKEL can be improved by applying some stronger ECC, but also show that the traditional Binary Relevance approach can be enhanced by learning more parity-checking labels. In addition, our study on different ECC helps understand the trade-off between the strength of ECC and the hardness of the base learning tasks. ER -
APA
Ferng, C. & Lin, H.. (2011). Multi-label Classification with Error-correcting Codes. Proceedings of the Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 20:281-295 Available from https://proceedings.mlr.press/v20/ferng11.html.

Related Material