A theoretical study on six classifier fusion strategies

IEEE Transactions on Pattern Analysis and Machine Intelligence

Volumn 24, Issue 2, 2002, Pages 281-286

  Kuncheva, Ludmila I ^a,b  

^a IEEE   (United Kingdom)

^b BANGOR UNIVERSITY   (United Kingdom)

Author keywords

Classifier combination; Fusion methods; Independent classifiers; Majority vote; Order statistics; Theoretical error

Indexed keywords

CLASSIFICATION ERROR; CLASSIFIER COMBINATION; CLASSIFIER FUSION METHODS; POSTERIOR PROBABILITY;

CLASSIFICATION (OF INFORMATION); ERRORS; MATHEMATICAL MODELS; NORMAL DISTRIBUTION; PROBABILITY; PROBABILITY DENSITY FUNCTION;

PATTERN RECOGNITION;

EID: 0036472946     PISSN: 01628828     EISSN: None     Source Type: Journal    
DOI: 10.1109/34.982906     Document Type: Article

References (14)

1. Experimental evaluation of expert fusion strategies
2. Combining predictors
3. On combining classifiers
4. Decision templates for multiple classifier fusion: An experimental comparison
5. Ten measures of diversity in classifier ensembles: Limits for two classifiers
6. Conceptual modeling of coincident failures in multiversion software
7. Theoretical views of boosting
8. Comparison between product and mean classifier combination rules
9. Error correlation and error reduction in ensemble classifiers
10. Linear and order statistics combiners for pattern classification
11. Combining classifiers for the Recognition of handwritten digits