Multilabel classification through random graph ensembles

DOI10.1007/S10994-014-5465-9zbMATH Open1331.68194arXiv1310.8428OpenAlexW2115386194MaRDI QIDQ2353005FDOQ2353005

Hongyu Su, Juho Rousu

Publication date: 7 July 2015

Published in: Machine Learning (Search for Journal in Brave)

Abstract: We present new methods for multilabel classification, relying on ensemble learning on a collection of random output graphs imposed on the multilabel and a kernel-based structured output learner as the base classifier. For ensemble learning, differences among the output graphs provide the required base classifier diversity and lead to improved performance in the increasing size of the ensemble. We study different methods of forming the ensemble prediction, including majority voting and two methods that perform inferences over the graph structures before or after combining the base models into the ensemble. We compare the methods against the state-of-the-art machine learning approaches on a set of heterogeneous multilabel benchmark problems, including multilabel AdaBoost, convex multitask feature learning, as well as single target learning approaches represented by Bagging and SVM. In our experiments, the random graph ensembles are very competitive and robust, ranking first or second on most of the datasets. Overall, our results show that random graph ensembles are viable alternatives to flat multilabel and multitask learners.

Full work available at URL: https://arxiv.org/abs/1310.8428

Recommendations

• Improving multilabel classification via heterogeneous ensemble methods
• Ensembles of relational classifiers
• An ensemble method for high-dimensional multilabel data
• Multi-label feature ranking with ensemble methods
• Ensemble learning for relational data

zbMATH Keywords

graphical modelskernel methodsensemble methodsmultilabel classificationstructured output

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Cites Work

• A decision-theoretic generalization of on-line learning and an application to boosting
• Bagging predictors
• Title not available (Why is that?)
• An introduction to support vector machines and other kernel-based learning methods.
• Improved boosting algorithms using confidence-rated predictions
• Multiple Comparisons Among Means
• BoosTexter: A boosting-based system for text categorization
• Convex multi-task feature learning
• MAP Estimation Via Agreement on Trees: Message-Passing and Linear Programming
• Title not available (Why is that?)
• Well-posedness of the Cauchy problem for the Hirota equation in Sobolev spaces \(H^s\)
• Multilabel classification through random graph ensembles

Cited In (3)

• A multilevel approach for learning from labeled and unlabeled data on graphs
• Multilabel classification through random graph ensembles
• Scalable multi-output label prediction: from classifier chains to classifier trellises

Uses Software

• AdaBoost.MH
• BoosTexter

This page was built for publication: Multilabel classification through random graph ensembles

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2353005)