Information theoretic combination of pattern classifiers

DOI10.1016/j.patcog.2010.04.013zbMath1344.68197MaRDI QIDQ991948

Jean-Philippe Thiran, Julien Meynet

Publication date: 8 September 2010

Published in: Pattern Recognition (Search for Journal in Brave)

Full work available at URL: http://infoscience.epfl.ch/record/149293

zbMATH Keywords

pattern recognition; mutual information; information theory; machine learning; diversity; classifier combination

Mathematics Subject Classification ID

68T05: Learning and adaptive systems in artificial intelligence

68T10: Pattern recognition, speech recognition

94A15: Information theory (general)

Related Items

A new methodology for generating and combining statistical forecasting models to enhance competitive event prediction, Information theoretic combination of pattern classifiers, A probabilistic model of classifier competence for dynamic ensemble selection, A novel method for combining Bayesian networks, theoretical analysis, and its applications, Protein subcellular localization in human and hamster cell lines: employing local ternary patterns of fluorescence microscopy images

Uses Software

• UCI-ml
• PRTools
• AdaBoost.MH

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Bagging predictors
• An analysis of diversity measures
• From error probability to information theoretic (multi-modal) signal processing
• Information theoretic combination of pattern classifiers
• A decision-theoretic generalization of on-line learning and an application to boosting
• Measures of diversity in classifier ensembles and their relationship with the ensemble accuracy
• Limits on the majority vote accuracy in classifier fusion
• Lower and upper bounds for misclassification probability based on Renyi's information
• Combining Pattern Classifiers
• Learning from examples with information theoretic criteria
• Random forests