Threshold optimisation for multi-label classifiers

DOI10.1016/J.PATCOG.2013.01.012MaRDI QIDQ743256zbMATH OpenOpenAlexFDO

Authors Ignazio Pillai, Giorgio Fumera, Fabio Roli

Publication date 23 September 2014

Published in Pattern Recognition (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.patcog.2013.01.012

zbMATH Keywords

\(F\) measuremulti-label classificationprecision-recall curveS-cut thresholding

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Pattern recognition, speech recognition (68T10)

Recommendations

• Designing multi-label classifiers that maximize F measures: state of the art
• On the Bayes-optimality of F-measure maximizers
• Multilabel classifiers with a probabilistic thresholding strategy
• On reoptimizing multi-class classifiers
• Optimal ranking in multi-label classification using local precision rates

Cites work

• A non-classical logic for information retrieval
• scientific article; zbMATH DE number 1717316 (Why is no real title available?)
• Introduction to Information Retrieval

Cited in

(10)

• On reoptimizing multi-class classifiers
• Efficient optimization of \(F\)-measure with cost-sensitive SVM
• Optimal rates for nonparametric F-score binary classification via post-processing
• Designing multi-label classifiers that maximize F measures: state of the art
• Efficient \(F\) measure maximization via weighted maximum likelihood
• Multi-label core vector machine with a zero label
• On the Bayes-optimality of F-measure maximizers
• A stacking weighted \(k\)-nearest neighbour with thresholding
• Multilabel classifiers with a probabilistic thresholding strategy
• Natural-neighborhood based, label-specific undersampling for imbalanced, multi-label data

Describes a project that uses

Uses Software

• LIBSVM
• MULAN

This page was built for publication: Threshold optimisation for multi-label classifiers

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q743256)