bibtype	C - Conference Paper (international conference)
ARLID	0311211
utime	20240103190335.8
mtime	20090326235959.9
WOS	000259567200006
title (primary) (eng)	Extraction of Binary Features by Probabilistic Neural Networks
specification	page_count 10 s.
serial	ARLID cav_un_epca*0311333 ISBN 978-3-540-87558-1 title Artificial Neural Networks - ICANN 2008 part_num Part II page_num 52-61 publisher place Berlin name Springer year 2008 editor name1 Kůrková name2 V. editor name1 Neruda name2 R. editor name1 Koutník name2 J.
title (cze)	Extrakce binárních příznaků pomocí pravděpodobnostních neuronových sítí
keyword	Probabilistic neural networks
keyword	Feature extraction
keyword	Recognition of numerals
author (primary)	ARLID cav_un_auth*0101091 name1 Grim name2 Jiří institution UTIA-B full_dept Department of Pattern Recognition fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
cas_special	project project_id 1M0572 agency GA MŠk ARLID cav_un_auth*0001814 project project_id 2C06019 agency GA MŠk country CZ ARLID cav_un_auth*0216518 project project_id GA102/07/1594 agency GA ČR ARLID cav_un_auth*0228611 research CEZ:AV0Z10750506 abstract (eng) In order to design probabilistic neural networks in the framework of pattern recognition we estimate class-conditional probability distributions in the form of finite mixtures of product components. As the mixture components correspond to neurons we specify the properties of neurons in terms of component parameters. The probabilistic features defined by neuron outputs can be used to transform the classification problem without information loss and, simultaneously, the Shannon entropy of the feature space is minimized. We show that, instead of dimensionality reduction, the decision problem can be simplified by using binary approximation of the probabilistic features. In experiments the resulting binary features improve recognition accuracy but also they are nearly independent - in accordance with the minimum entropy property. abstract (cze) Extrakce binárních příznaků pomocí pravděpodobnostních neuronových sítí action ARLID cav_un_auth*0241921 name ICANN 2008. International Conference on Artificial Neural Networks /18./ place Prague dates 03.09.2008-06.09.2008 country CZ reportyear 2009 RIV IN permalink http://hdl.handle.net/11104/0162890 arlyear 2008 mrcbU34 000259567200006 WOS mrcbU63 cav_un_epca*0311333 Artificial Neural Networks - ICANN 2008 Part II 978-3-540-87558-1 52 61 Berlin Springer 2008 Lecture Notes in Computer Science 5164 mrcbU67 Kůrková V. 340 mrcbU67 Neruda R. 340 mrcbU67 Koutník J. 340