Condensed Filter Tree for Cost-Sensitive Multi-Label Classification

Chun-Liang Li, Hsuan-Tien Lin
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(1):423-431, 2014.

Abstract

Different real-world applications of multi-label classification often demand different evaluation criteria. We formalize this demand with a general setup, cost-sensitive multi-label classification (CSMLC), which takes the evaluation criteria into account during learning. Nevertheless, most existing algorithms can only focus on optimizing a few specific evaluation criteria, and cannot systematically deal with different ones. In this paper, we propose a novel algorithm, called condensed filter tree (CFT), for optimizing any criteria in CSMLC. CFT is derived from reducing CSMLC to the famous filter tree algorithm for cost-sensitive multi-class classification via constructing the label powerset. We successfully cope with the difficulty of having exponentially many extended-classes within the powerset for representation, training and prediction by carefully designing the tree structure and focusing on the key nodes. Experimental results across many real-world datasets validate that CFT is competitive with special purpose algorithms on special criteria and reaches better performance on general criteria.

Cite this Paper

BibTeX
@InProceedings{pmlr-v32-lia14, title = {Condensed Filter Tree for Cost-Sensitive Multi-Label Classification}, author = {Li, Chun-Liang and Lin, Hsuan-Tien}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {423--431}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {1}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/lia14.pdf}, url = {https://proceedings.mlr.press/v32/lia14.html}, abstract = {Different real-world applications of multi-label classification often demand different evaluation criteria. We formalize this demand with a general setup, cost-sensitive multi-label classification (CSMLC), which takes the evaluation criteria into account during learning. Nevertheless, most existing algorithms can only focus on optimizing a few specific evaluation criteria, and cannot systematically deal with different ones. In this paper, we propose a novel algorithm, called condensed filter tree (CFT), for optimizing any criteria in CSMLC. CFT is derived from reducing CSMLC to the famous filter tree algorithm for cost-sensitive multi-class classification via constructing the label powerset. We successfully cope with the difficulty of having exponentially many extended-classes within the powerset for representation, training and prediction by carefully designing the tree structure and focusing on the key nodes. Experimental results across many real-world datasets validate that CFT is competitive with special purpose algorithms on special criteria and reaches better performance on general criteria.} }
Endnote
%0 Conference Paper %T Condensed Filter Tree for Cost-Sensitive Multi-Label Classification %A Chun-Liang Li %A Hsuan-Tien Lin %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-lia14 %I PMLR %P 423--431 %U https://proceedings.mlr.press/v32/lia14.html %V 32 %N 1 %X Different real-world applications of multi-label classification often demand different evaluation criteria. We formalize this demand with a general setup, cost-sensitive multi-label classification (CSMLC), which takes the evaluation criteria into account during learning. Nevertheless, most existing algorithms can only focus on optimizing a few specific evaluation criteria, and cannot systematically deal with different ones. In this paper, we propose a novel algorithm, called condensed filter tree (CFT), for optimizing any criteria in CSMLC. CFT is derived from reducing CSMLC to the famous filter tree algorithm for cost-sensitive multi-class classification via constructing the label powerset. We successfully cope with the difficulty of having exponentially many extended-classes within the powerset for representation, training and prediction by carefully designing the tree structure and focusing on the key nodes. Experimental results across many real-world datasets validate that CFT is competitive with special purpose algorithms on special criteria and reaches better performance on general criteria.
RIS
TY - CPAPER TI - Condensed Filter Tree for Cost-Sensitive Multi-Label Classification AU - Chun-Liang Li AU - Hsuan-Tien Lin BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/01/27 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-lia14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 1 SP - 423 EP - 431 L1 - http://proceedings.mlr.press/v32/lia14.pdf UR - https://proceedings.mlr.press/v32/lia14.html AB - Different real-world applications of multi-label classification often demand different evaluation criteria. We formalize this demand with a general setup, cost-sensitive multi-label classification (CSMLC), which takes the evaluation criteria into account during learning. Nevertheless, most existing algorithms can only focus on optimizing a few specific evaluation criteria, and cannot systematically deal with different ones. In this paper, we propose a novel algorithm, called condensed filter tree (CFT), for optimizing any criteria in CSMLC. CFT is derived from reducing CSMLC to the famous filter tree algorithm for cost-sensitive multi-class classification via constructing the label powerset. We successfully cope with the difficulty of having exponentially many extended-classes within the powerset for representation, training and prediction by carefully designing the tree structure and focusing on the key nodes. Experimental results across many real-world datasets validate that CFT is competitive with special purpose algorithms on special criteria and reaches better performance on general criteria. ER -
APA
Li, C. & Lin, H.. (2014). Condensed Filter Tree for Cost-Sensitive Multi-Label Classification. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(1):423-431 Available from https://proceedings.mlr.press/v32/lia14.html.

Related Material