On output independence and complementariness in rank-based multiple classifier decision systems

Pattern Recognition

Volumn 34, Issue 12, 2001, Pages 2319-2330

  Saranli, Afşar ^a   Demirekler, Mübeccel ^a  

^a MIDDLE EAST TECHNICAL UNIVERSITY   (Turkey)

Author keywords

Bayesian formalism; Classifier observation space; Complementariness; Entropy; Event space partitioning; Independence; Multiple classifier systems; Mutual information; Ranks; Statistical classifier combination; Statistical decision combination; Statistical pattern recognition

Indexed keywords

COMPUTER SIMULATION; DECISION SUPPORT SYSTEMS; INFORMATION THEORY; SPACE APPLICATIONS;

OUTPUT INDEPENDENCE;

PATTERN RECOGNITION;

IMAGE CLASSIFICATION; PATTERN RECOGNITION; THEORETICAL STUDY;

EID: 0035546359     PISSN: 00313203     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0031-3203(00)00175-8     Document Type: Article

References (30)

1. Decision combination in multiple classifier systems
2. Methods of combining multiple classifiers and their applications to handwriting recognition
3. A method of combining multiple experts for the recognition of unconstrained handwritten numerals
4. Recognition of handwritten digits by combining independent learning vector quantizations
5. Democracy in neural nets: Voting schemes for classification
6. Combining the results of several neural network classifiers
7. On combining classifiers
8. Handwritten numerical recognition based on multiple algorithms
9. Data fusion techniques in speaker recognition
10. Neural networks for discrimination and modelization of speakers
11. A study on improving decisions in closed set speaker identification
12. An approach to speaker identification using multiple classifiers
13. Combination of face classifiers for person identification
14. Person identification using multiple cues
15. Stacked generalization
16. Parallel consensual neural networks
17. Consensus theoretic classification methods
18. Decision fusion
19. Integration of neural networks and decision tree classifiers for automated cytology screening
20. A patient-adaptable ECG beat classifier using a mixture of experts approach
21. Analysis of decision boundaries in linearly combined neural classifiers
22. When networks disagree: Ensemble methods for hybrid neural networks
23. Learning ranks with neural networks
24. Lowering variance of decisions using artificial neural networks portfolios
25. Neural network ensembles
26. Estimating the Bayes error rate through classifier combining
27. A statistical unified framework for rank-based multiple classifier decision combination
28. Rank-based multiple classifier decision combination: A theoretical study