Circular Chain Classifiers

Jesús Joel Rivas, Felipe Orihuela-Espina, Luis Enrique Succar
Proceedings of the Ninth International Conference on Probabilistic Graphical Models, PMLR 72:380-391, 2018.

Abstract

Chain Classifiers (CC) are an alternative for multi-label classification that is efficient and provides, in general, good results. However, it is not clear how to define the order of the chain. Different orders tend to produce different outcomes. We propose an extension to chain classifiers called “Circular Chain Classifiers" (CCC), in which the propagation of the classes of the previous binary classifiers is done iteratively in a circular way. After the first cycle, the predictions from the base classifiers are entered as additional attributes to the first one in the chain. This process continues for all the classifiers in the chain, and it is repeated for a prefixed number of cycles or until convergence. Using two datasets, we empirically established that CCC: (i) converges in few iterations (in general, 3 or 4), (ii) the initial order of the chain does not have a significant impact on the results. CCC performance was also compared against binary relevance and chain classifiers producing statistically superior results. The main contribution of CCC is its independence from the preestablished order of the chain, outperforming CC.

Cite this Paper


BibTeX
@InProceedings{pmlr-v72-rivas18a, title = {Circular Chain Classifiers}, author = {Rivas, Jes\'{u}s Joel and Orihuela-Espina, Felipe and Succar, Luis Enrique}, booktitle = {Proceedings of the Ninth International Conference on Probabilistic Graphical Models}, pages = {380--391}, year = {2018}, editor = {Kratochvíl, Václav and Studený, Milan}, volume = {72}, series = {Proceedings of Machine Learning Research}, month = {11--14 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v72/rivas18a/rivas18a.pdf}, url = {https://proceedings.mlr.press/v72/rivas18a.html}, abstract = {Chain Classifiers (CC) are an alternative for multi-label classification that is efficient and provides, in general, good results. However, it is not clear how to define the order of the chain. Different orders tend to produce different outcomes. We propose an extension to chain classifiers called “Circular Chain Classifiers" (CCC), in which the propagation of the classes of the previous binary classifiers is done iteratively in a circular way. After the first cycle, the predictions from the base classifiers are entered as additional attributes to the first one in the chain. This process continues for all the classifiers in the chain, and it is repeated for a prefixed number of cycles or until convergence. Using two datasets, we empirically established that CCC: (i) converges in few iterations (in general, 3 or 4), (ii) the initial order of the chain does not have a significant impact on the results. CCC performance was also compared against binary relevance and chain classifiers producing statistically superior results. The main contribution of CCC is its independence from the preestablished order of the chain, outperforming CC.} }
Endnote
%0 Conference Paper %T Circular Chain Classifiers %A Jesús Joel Rivas %A Felipe Orihuela-Espina %A Luis Enrique Succar %B Proceedings of the Ninth International Conference on Probabilistic Graphical Models %C Proceedings of Machine Learning Research %D 2018 %E Václav Kratochvíl %E Milan Studený %F pmlr-v72-rivas18a %I PMLR %P 380--391 %U https://proceedings.mlr.press/v72/rivas18a.html %V 72 %X Chain Classifiers (CC) are an alternative for multi-label classification that is efficient and provides, in general, good results. However, it is not clear how to define the order of the chain. Different orders tend to produce different outcomes. We propose an extension to chain classifiers called “Circular Chain Classifiers" (CCC), in which the propagation of the classes of the previous binary classifiers is done iteratively in a circular way. After the first cycle, the predictions from the base classifiers are entered as additional attributes to the first one in the chain. This process continues for all the classifiers in the chain, and it is repeated for a prefixed number of cycles or until convergence. Using two datasets, we empirically established that CCC: (i) converges in few iterations (in general, 3 or 4), (ii) the initial order of the chain does not have a significant impact on the results. CCC performance was also compared against binary relevance and chain classifiers producing statistically superior results. The main contribution of CCC is its independence from the preestablished order of the chain, outperforming CC.
APA
Rivas, J.J., Orihuela-Espina, F. & Succar, L.E.. (2018). Circular Chain Classifiers. Proceedings of the Ninth International Conference on Probabilistic Graphical Models, in Proceedings of Machine Learning Research 72:380-391 Available from https://proceedings.mlr.press/v72/rivas18a.html.

Related Material