TU Darmstadt / ULB / TUbiblio

Exploiting Anti-monotonicity of Multi-label Evaluation Measures for Inducing Multi-label Rules

Rapp, Michael and Loza Mencía, Johannes
Phung, Dinh and Tseng, Vincent S. and Webb, Geoffrey I. and Ho, Bao and Ganji, Mohadeseh and Rashidi, Lida (eds.) (2018):
Exploiting Anti-monotonicity of Multi-label Evaluation Measures for Inducing Multi-label Rules.
Cham, Springer International Publishing, In: Advances in Knowledge Discovery and Data Mining, Cham, ISBN 978-3-319-93034-3,
[Conference or Workshop Item]

Abstract

Exploiting dependencies between labels is considered to be crucial for multi-label classification. Rules are able to expose label dependencies such as implications, subsumptions or exclusions in a human-comprehensible and interpretable manner. However, the induction of rules with multiple labels in the head is particularly challenging, as the number of label combinations which must be taken into account for each rule grows exponentially with the number of available labels. To overcome this limitation, algorithms for exhaustive rule mining typically use properties such as anti-monotonicity or decomposability in order to prune the search space. In the present paper, we examine whether commonly used multi-label evaluation metrics satisfy these properties and therefore are suited to prune the search space for multi-label heads.

Item Type: Conference or Workshop Item
Erschienen: 2018
Editors: Phung, Dinh and Tseng, Vincent S. and Webb, Geoffrey I. and Ho, Bao and Ganji, Mohadeseh and Rashidi, Lida
Creators: Rapp, Michael and Loza Mencía, Johannes
Title: Exploiting Anti-monotonicity of Multi-label Evaluation Measures for Inducing Multi-label Rules
Language: German
Abstract:

Exploiting dependencies between labels is considered to be crucial for multi-label classification. Rules are able to expose label dependencies such as implications, subsumptions or exclusions in a human-comprehensible and interpretable manner. However, the induction of rules with multiple labels in the head is particularly challenging, as the number of label combinations which must be taken into account for each rule grows exponentially with the number of available labels. To overcome this limitation, algorithms for exhaustive rule mining typically use properties such as anti-monotonicity or decomposability in order to prune the search space. In the present paper, we examine whether commonly used multi-label evaluation metrics satisfy these properties and therefore are suited to prune the search space for multi-label heads.

Place of Publication: Cham
Publisher: Springer International Publishing
ISBN: 978-3-319-93034-3
Divisions: DFG-Graduiertenkollegs
DFG-Graduiertenkollegs > Research Training Group 1994 Adaptive Preparation of Information from Heterogeneous Sources
Event Title: Advances in Knowledge Discovery and Data Mining
Event Location: Cham
Date Deposited: 23 Oct 2018 13:56
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item