Multi-label Classification Using Ensembles of Pruned Sets
Authors:
- Jesse ReadJRJesse Read
- Bernhard PfahringerBPBernhard Pfahringer
- Geoff HolmesGHGeoff Holmes
- 1 December 2008
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15504786,p. 995-1000
Abstract
This paper presents a pruned sets method (PS) for multi-label classification. It is centred on the concept of treating sets of labels as single labels. This allows the classification process to inherently take into account correlations between labels. By pruning these sets, PS focuses only on the most important correlations, which reduces complexity and improves accuracy. By combining pruned sets in an ensemble scheme (EPS), new label sets can be formed to adapt to irregular or complex data. The results from experimental evaluation on a variety of multi-label datasets show that [E]PS can achieve better performance and train much faster than other multi-label methods.Keywords
This publication has 5 references indexed in Scilit:
- Decision trees for hierarchical multi-label classificationMachine Learning, 2008
- Random k-Labelsets: An Ensemble Method for Multilabel Classification2007
- Multi-Label ClassificationInternational Journal of Data Warehousing and Mining, 2007
- Discriminative Methods for Multi-labeled Classification2004
- BoosTexter: A Boosting-based System for Text CategorizationMachine Learning, 2000