We propose a generalization of multilabel classification that we refer to as graded multilabel classification. The key idea is that, instead of requesting a yes-no answer to the question of class membership or, say, relevance of a class label for an instance, we allow for a graded membership of an instance, measured on an ordinal scale of membership degrees. This extension is motivated by practical applications in which a graded or partial class membership is natural. Apart from introducing the basic setting, we propose two general strategies for reducing graded multilabel problems to conventional (multilabel) classification problems. Moreover, we address the question of how to extend performance metrics commonly used in multilabel classification to the graded setting, and present first experimental results.
%0 Conference Paper
%1 kdml18
%A Cheng, Weiwei
%A Dembczynski, Krzysztof
%A Hüllermeier., Eyke
%B Proceedings of LWA2010 - Workshop-Woche: Lernen, Wissen & Adaptivitaet
%C Kassel, Germany
%D 2010
%E Atzmüller, Martin
%E Benz, Dominik
%E Hotho, Andreas
%E Stumme, Gerd
%K classification degrees framework fuzzy graded membership multilabel ordinal reduction room:0446 session:joint3 set theory workshop:kdml
%T Graded Multilabel Classification: The Ordinal Case
%U http://www.kde.cs.uni-kassel.de/conf/lwa10/papers/kdml18.pdf
%X We propose a generalization of multilabel classification that we refer to as graded multilabel classification. The key idea is that, instead of requesting a yes-no answer to the question of class membership or, say, relevance of a class label for an instance, we allow for a graded membership of an instance, measured on an ordinal scale of membership degrees. This extension is motivated by practical applications in which a graded or partial class membership is natural. Apart from introducing the basic setting, we propose two general strategies for reducing graded multilabel problems to conventional (multilabel) classification problems. Moreover, we address the question of how to extend performance metrics commonly used in multilabel classification to the graded setting, and present first experimental results.
@inproceedings{kdml18,
abstract = {We propose a generalization of multilabel classification that we refer to as graded multilabel classification. The key idea is that, instead of requesting a yes-no answer to the question of class membership or, say, relevance of a class label for an instance, we allow for a graded membership of an instance, measured on an ordinal scale of membership degrees. This extension is motivated by practical applications in which a graded or partial class membership is natural. Apart from introducing the basic setting, we propose two general strategies for reducing graded multilabel problems to conventional (multilabel) classification problems. Moreover, we address the question of how to extend performance metrics commonly used in multilabel classification to the graded setting, and present first experimental results.},
added-at = {2010-10-05T14:15:12.000+0200},
address = {Kassel, Germany},
author = {Cheng, Weiwei and Dembczynski, Krzysztof and Hüllermeier., Eyke},
biburl = {https://www.bibsonomy.org/bibtex/2d2bb414889b1f18086ecbfd86a4a8b15/lwa2010},
booktitle = {Proceedings of LWA2010 - Workshop-Woche: Lernen, Wissen {\&} Adaptivitaet},
crossref = {lwa2010},
editor = {Atzmüller, Martin and Benz, Dominik and Hotho, Andreas and Stumme, Gerd},
interhash = {1be734581900bed79c4f76f56097446c},
intrahash = {d2bb414889b1f18086ecbfd86a4a8b15},
keywords = {classification degrees framework fuzzy graded membership multilabel ordinal reduction room:0446 session:joint3 set theory workshop:kdml},
presentation_end = {2010-10-05 12:30:00},
presentation_start = {2010-10-05 12:00:00},
room = {0446},
session = {joint3},
timestamp = {2010-10-05T14:15:14.000+0200},
title = {Graded Multilabel Classification: The Ordinal Case},
track = {kdml},
url = {http://www.kde.cs.uni-kassel.de/conf/lwa10/papers/kdml18.pdf},
year = 2010
}