Title :
Using 2-additive fuzzy measure in Multiple Classifier System
Author :
Zhao, Li ; Chen, Ai-xia ; Li, Ning ; Yuan, Guo-qiang ; Zhang, Guo-fang
Author_Institution :
Key Lab. of Machine Learning & Comput. Intell., Hebei Univ., Baoding, China
Abstract :
Fuzzy measure and integral are widely used in multiple classifier system (MCS). But the number of coefficients involved in the fuzzy integral model grows exponentially with the number of classifiers to be aggregated. The main difficulty is to identify all these coefficients. This paper does an attempt Using 2-additive fuzzy measure in multiple classifier system. Our conclusion is that when different interactions exist in different classifiers the complexity of the computation can be significantly reduced by 2-order additive measure.A simple example is included to illustrate the 2-order additive measure.
Keywords :
fuzzy set theory; integral equations; pattern classification; 2-additive fuzzy measure; fuzzy integral model; multiple classifier system; Computational intelligence; Computer science; Cybernetics; Educational institutions; Electronic mail; Finance; Fuzzy sets; Fuzzy systems; Machine learning; Mathematics; λ - fuzzy measure; 2-additive fuzzy measure; Fuzzy measure; Interaction; Multiple Classifier System;
Conference_Titel :
Machine Learning and Cybernetics, 2009 International Conference on
Conference_Location :
Baoding
Print_ISBN :
978-1-4244-3702-3
Electronic_ISBN :
978-1-4244-3703-0
DOI :
10.1109/ICMLC.2009.5212369
