JACIII Vol.5 No.1 pp. 37-43
doi: 10.20965/jaciii.2001.p0037


Rank-Based Multiple Classifier Decision Combination: A Theoretical Study

Afsar Saranli and Mubeccel Demirekler

Speech Processing Laboratory, Department of Electrical and Electronics Engineering Middle East Technical University Ismet Inonu Bulv. 06531, Ankara, Turkiye

October 12, 2000
December 5, 2000
January 20, 2001
Decision combination,multiple classifiet systems, rank-based classifiers
This study presents a theoretical investigation of the rank-based multiple classifier decision problem for closed-set pattern identification. The problem of combining the decisions of more than one classifiers with raw outputs in the form of candidate class rankings is considered and formulated as a general discrete optimization problem with an objective function based on the total probability of correct decision. This formulation uses certain performance statistics about the joint behavior of the ensemble of classifiers, which need to be estimated from cross-validation data. An initial approach leads to an integer (binary) programming problem with a simple and global optimum solution but of prohibitive dimensionality. Therefore, we present a partitioning formalism under which this dimensionality can be reduced by incorporating our prior knowledge about the problem domain and the structure of the training data. It is also shown that formalism can effectively explain a number of successfully used combination approaches in the literature.
Cite this article as:
A. Saranli and M. Demirekler, “Rank-Based Multiple Classifier Decision Combination: A Theoretical Study,” J. Adv. Comput. Intell. Intell. Inform., Vol.5 No.1, pp. 37-43, 2001.
Data files:

*This site is desgined based on HTML5 and CSS3 for modern browsers, e.g. Chrome, Firefox, Safari, Edge, Opera.

Last updated on Jul. 19, 2024