Combining multiple classifiers in probabilistic neural networks

0410332 20240103182206.620060210235959.9 3-540-67704-6 Combining multiple classifiers in probabilistic neural networks Berlin Springer 2000 10 s. Lecture Notes in Computer Science. 1857 Multiple Classifier Systems157-166KittlerJ.RoliF. cav_un_auth*0101091 Grim Jiří UTIA-B Department of Pattern Recognition Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0021087 Kittler J. GB cav_un_auth*0101182 Pudil Pavel UTIA-B Department of Pattern Recognition Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101197 Somol Petr UTIA-B Department of Pattern Recognition Ústav teorie informace a automatizace AV ČR, v. v. i. 12B 09K IAA2075703 GA AV CZ cav_un_auth*0218731 VS96063 MŠMT CZ cav_un_auth*0025066 KSK1075601 GA AV CZ cav_un_auth*0027435 AV0Z1075907 The paper summarizes main features of a new probabilistic approach to neural networks in the framework of statistical pattern recognition. Assuming approximation of class-conditional distributions by finite mixtures we identify formal neurons with the components of finite mixtures and therefore the EM algorithm can be used to optimize the parameters of neurons. In order to prevent the arising information loss we propose a parallel use of the output variables to design the Bayesian classifier. cav_un_auth*0212643 First International Workshop MCS 2000 Cagliari IT 21.06.2000-23.06.2000 BB RO http://hdl.handle.net/11104/0130422 UTIA-B 20000048 2000 2000 Berlin Springer 3-540-67704-6 Multiple Classifier Systems 157 166 Kittler J. 340 Roli F. 340