Abstract

When more than a single classifier has been trained for the same recognition problem the question arises how this set of classifiers may be combined into a final decision rule. Several fixed combining rules are used that depend on the output values of the base classifiers only. They are almost always suboptimal.

Citations

1998	Bagging predictors - Breiman - 1996
1714	Schapire R: A decision-theoretic generalization of online learning and an application to boosting - Freund - 1997
749	On combining classifiers - Kittler - 1998
487	Statistical pattern recognition: A review - Jain, Duin, et al.
248	Decision Combination in Multiple Classifier Systems - Ho, Hull, et al. - 1994
247	TK: The Random Subspace Method for Constructing Decision Forests - Ho - 1998
127	Learning Machines - Nilsson
107	A Method of Combining Multiple Experts for the Recognition of Unconstrained Handwritten Numerals - Huang, Suen - 1995
98	Fuzzy Models for Pattern Recognition - Bezdek, Pal - 1992
77	2001 Decision Templates for Multiple Classifier Fusion: An Experimental Comparison, Pattern Recognition - Kuncheva, Bezdek, et al. - 2001
62	A theoretical study on six classifier fusion strategies - Kuncheva - 2002
39	Multiple classifier systems - Haindl, Kittler, et al.
35	D.M.J.: Experiments with classifier combining rules - Duin, Tax - 2000
32	Dissimilarity representations allow for building good classifiers - Pekalska, Duin - 2002
30	Fuzzy Classifier Design - Kuncheva - 2000
25	Methods of Combining Multiple Classifiers with Different Features and Their Applications to Text-Independent Speaker Identification - Chen, Wang, et al. - 1997
25	Switching between selection and fusion in combining classifiers: An experiment - Kuncheva - 2002
20	F.: Dynamic classifier selection based on multiple classifier behaviour. Pattern Recognition 34 - Giacinto, Roli - 2001
17	An approach to automatic design of multiple classifier systems - Giacinto, Roli - 2001
13	Classifier conditional posterior probabilities - Duin, Tax - 1998
12	Automatic pattern recognition by similarity representations - Pekalska, Duin - 2001
6	An experimental comparison of fixed and trained fusion rules for crisp classifiers outputs - Roli, Raudys, et al. - 2002
2	Measures of diversity in classifier ensembles. submitted - Kuncheva, Whitaker - 2002
1	Reduction a Boasting Bias of Linear Experts - Raudys, Janeliunas - 2002