Title :
Strategies for weighted combination of classifiers employing shared and distinct representations
Author :
Hojjatoleslami, A. ; Kittler, J.
Author_Institution :
Sch. of Electron. Eng., Inf. Technol. & Math., Surrey Univ., Guildford, UK
Abstract :
We present a theoretical framework for combining soft decision outputs of multiple experts employing mixed (some shared and some distinct) representations of patterns to be classified. Using this framework combination strategies are developed and their error sensitivity studied We show that nonuniform weighting combination strategies can be derived by taking into account the confidence of experts in their output
Keywords :
pattern classification; error sensitivity; expert confidence; multiple experts; nonuniform weighting combination strategies; pattern representations; soft decision outputs; weighted classifier combination; Bayesian methods; Councils; Information technology; Mathematics; Pattern recognition; Performance analysis; Probability density function; Signal processing; Speech processing;
Conference_Titel :
Pattern Recognition, 1998. Proceedings. Fourteenth International Conference on
Conference_Location :
Brisbane, Qld.
Print_ISBN :
0-8186-8512-3
DOI :
10.1109/ICPR.1998.711149
