A framework for multi-label classification extended by Error Correcting Output Codes (ECOCs) is introduced and empirically examined in the article. The solution assumes the base multi-label classifiers to be a noisy channel and applies ECOCs in order to recover the classification errors made by individual classifiers. ; The framework was examined through exhaustive studies over combinations of three distinct classification algorithms and four ECOC methods employed in the multi-label classification problem. ; The experimental results revealed that (i) the Bode?Chaudhuri?Hocquenghem (BCH) code matched with any multi-label classifier results in better classification quality; (ii) the accuracy of the binary relevance classification method strongly depends on the coding scheme; (iii) the label power-set and the RAkEL classifier consume the same time for computation irrespective of the coding utilized; (iv) in general, they are not suitable for ECOCs because they are not capable to benefit from ECOC correcting abilities; (v) the all-pairs code combined with binary relevance is not suitable for datasets with larger label sets.
Oct 14, 2021
Sep 15, 2021
|Multi-label classification using error correcting output codes||Oct 14, 2021|
Nowicki, Adam Grochowski, Michał Duzinkiewicz, Kazimierz Cordón, Oskar - ed. Kazienko, Przemysław - ed.
Czarnowski, Ireneusz Jędrzejowicz, Piotr Korbicz, Józef - red. Uciński, Dariusz - red.
Troć, Maciej Unold, Olgierd Korbicz, Józef - red. Uciński, Dariusz - red.
Kasiński, Andrzej Ponulak, Filip Korbicz, Józef - red.
Sumi, Monira S. Zaman, Faisal M. Hirose, Hideo Cordón, Oskar - ed. Kazienko, Przemysław - ed.
Trawiński, Bogdan Smętek, Magdalena Telec, Zbigniew Lasota, Tadeusz Cordón, Oskar - ed. Kazienko, Przemysław - ed.
Kładoczny, Piotr Sztyber, Radosław - red. nacz.
Ćwiąkała, Michał Greinert, Andrzej - red. Kołodziejczyk, Urszula - red.