Declaratively Capturing Local Label Correlations with Multi-Label Trees

Reem M Al-Otaibi, Meelis Kull, Peter A Flach

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

1 Citation (Scopus)
203 Downloads (Pure)

Abstract

The goal of multi-label classification is to predict multiple labels per data point simultaneously. Real-world applications tend to have high-dimensional label spaces, employing hundreds or even thousands of labels. While these labels could be predicted separately, by capturing label correlation we might achieve better predictive performance. In contrast with previous attempts in the literature that have modelled label correlations globally, this paper proposes a novel algorithm to model correlations and cluster labels locally. LaCovaC is a multi-label decision tree classifier that clusters labels into several dependent subsets at various points during training. The clusters are obtained locally by identifying the conditionally-dependent labels in localised regions of the feature space using the label correlation matrix. LaCovaC interleaves between two main decisions on the label matrix with training instances in rows and labels in columns: splitting this matrix vertically by partitioning the labels into subsets, or splitting it horizontally using features in the conventional way. Experiments on 13 benchmark datasets demonstrate that our proposal achieves competitive performance over a wide range of evaluation metrics when compared with the state-of-the-art multi-label classifiers.
Original languageEnglish
Title of host publicationProceedings of the 22nd European Conference on Artificial Intelligence (ECAI-2016), Including Prestigious Applications of Intelligent Systems (PAIS-2016)
EditorsGal A Kaminka, Maria Fox, Paolo Bouquet, Eyke Hüllermeier, Virginia Dignum, Frank Dignum, Frank van Harmelen
PublisherIOS Press
Pages1467-1475
Number of pages9
ISBN (Electronic)9781614996729
ISBN (Print)9781614996712
DOIs
Publication statusPublished - Aug 2016

Publication series

NameFrontiers in Artificial Intelligence and Applications
PublisherIOS Press
Volume285
ISSN (Print)0922-6389
ISSN (Electronic)1879-8314

Structured keywords

  • Jean Golding

Fingerprint Dive into the research topics of 'Declaratively Capturing Local Label Correlations with Multi-Label Trees'. Together they form a unique fingerprint.

Cite this