A statistical unified framework for rank-based multiple classifier decision combination

Saranlı, Afşar; Demirekler, Mübeccel

doi:10.1016/s0031-3203(00)00022-4

Cited by 24 publications

(10 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The arguments of Fact 1 and 2 and Theorem 1 also hold for the Borda Count method establishing its links with the introduced theory. As discussed in [25], it is also possible to show that the Logistic Regression method introduced first by Ho can be explained as a special case of the POS theory.…”

Section: The Borda Count Methodsmentioning

confidence: 97%

Rank-Based Multiple Classifier Decision Combination: A Theoretical Study

Saranlı¹,

Demirekler

2001

J. Adv. Comput. Intell. Intell. Inform.

View full text Add to dashboard Cite

show abstract

Section: The Borda Count Methodsmentioning

confidence: 97%

Rank-Based Multiple Classifier Decision Combination: A Theoretical Study

Saranlı¹,

Demirekler

2001

J. Adv. Comput. Intell. Intell. Inform.

View full text Add to dashboard Cite

show abstract

“…Behavior-knowledge space combination methods [13] are also based on ranks. Saranli and Demirekler [14] provide additional references for rank-based combination methods.…”

Section: A Rank-based Combinationsmentioning

confidence: 99%

Use of Identification Trial Statistics for the Combination of Biometric Matchers

Tulyakov

Govindaraju

2008

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

Abstract-Combination functions typically used in biometric identification systems consider as input parameters only those matching scores which are related to a single person in order to derive a combined score for that person. We discuss how such methods can be extended to utilize the matching scores corresponding to all people. The proposed combination methods account for dependencies between scores output by any single participating matcher. Our experiments demonstrate the advantage of using such combination methods when dealing with a large number of classes, as is the case with biometric person identification systems. The experiments are performed on the National Institute of Standards and Technology BSSR1 dataset and the combination methods considered include the likelihood ratio, neural network, and weighted sum.

show abstract

“…Since this estimate is based on observed data, there is inherent sample variance associated with these estimates. When these estimates are averaged over several Bayesian classifiers, this variance is reduced [18]. Also, certain classifiers may have better performance against specific targets or in certain situations [20].…”

Section: Classifier and Roc Fusion 241 Classifier Fusion Rulesmentioning

confidence: 99%

<title>Consistency results for the ROC curves of fused classifiers</title>

Bjerkaas¹

2004

SPIE Proceedings

View full text Add to dashboard Cite

A statistical unified framework for rank-based multiple classifier decision combination

Cited by 24 publications

References 33 publications

Rank-Based Multiple Classifier Decision Combination: A Theoretical Study

Rank-Based Multiple Classifier Decision Combination: A Theoretical Study

Use of Identification Trial Statistics for the Combination of Biometric Matchers

<title>Consistency results for the ROC curves of fused classifiers</title>

Contact Info

Product

Resources

About