https://mw.hh.se/caisr/index.php?action=history&feed=atom&title=Publications%3ASelecting_salient_features_for_classification_committees 2026-04-04T15:12:39Z Revision history for this page on the wiki MediaWiki 1.35.13 https://mw.hh.se/caisr/index.php?title=Publications:Selecting_salient_features_for_classification_committees&diff=3919&oldid=prev 2018-04-07T06:48:33Z

<p>Created page with &quot;&lt;div style=&#039;display: none&#039;&gt; == Do not edit this section == &lt;/div&gt; {{PublicationSetupTemplate|Author=Antanas Verikas, Marija Bacauskiene, Kerstin Malmqvist |PID=1195746 |Name=V...&quot;</p> <p><b>New page</b></p><div>&lt;div style=&#039;display: none&#039;&gt;<br /> == Do not edit this section ==<br /> &lt;/div&gt;<br /> {{PublicationSetupTemplate|Author=Antanas Verikas, Marija Bacauskiene, Kerstin Malmqvist<br /> |PID=1195746<br /> |Name=Verikas, Antanas (av) (0000-0003-2185-8973) (Högskolan i Halmstad (2804), Akademin för informationsteknologi (16904), Halmstad Embedded and Intelligent Systems Research (EIS) (3938), CAISR Centrum för tillämpade intelligenta system (IS-lab) (13650));Bacauskiene, Marija (Department of Applied Electronics, Kaunas University of Technology, Studentu 50, LT-3031, Kaunas, Lithuania);Malmqvist, Kerstin (kermal) (Högskolan i Halmstad (2804), Akademin för informationsteknologi (16904), Halmstad Embedded and Intelligent Systems Research (EIS) (3938))<br /> |Title=Selecting salient features for classification committees<br /> |PublicationType=Conference Paper<br /> |ContentType=Refereegranskat<br /> |Language=eng<br /> |Journal=<br /> |JournalISSN=<br /> |Status=<br /> |Volume=2714<br /> |Issue=<br /> |HostPublication=Artificial Neural Networks and Neural Information Processing — ICANN/ICONIP 2003<br /> |Conference=Joint International Conference on Artificial Neural Networks (ICANN)/International on Neural Information Processing (ICONIP), JUN 26-29, 2002, ISTANBUL, TURKEY<br /> |StartPage=35<br /> |EndPage=42<br /> |Year=2003<br /> |Edition=<br /> |Pages=<br /> |City=Heidelberg<br /> |Publisher=Springer Berlin/Heidelberg<br /> |Series=Lecture Notes in Computer Science ; 2714<br /> |SeriesISSN=0302-9743<br /> |ISBN=978-3-540-40408-8;978-3-540-44989-8<br /> |Urls=<br /> |ISRN=<br /> |DOI=http://dx.doi.org/10.1007/3-540-44989-2_5<br /> |ISI=000185378100005<br /> |PMID=<br /> |ScopusId=2-s2.0-21144434169<br /> |NBN=urn:nbn:se:hh:diva-35784<br /> |LocalId=<br /> |ArchiveNumber=<br /> |Keywords=Errors;Neural networks<br /> |Categories=Bioinformatik (beräkningsbiologi) (10203);Telekommunikation (20204);Bioinformatik och systembiologi (10610)<br /> |ResearchSubjects=<br /> |Projects=<br /> |Notes=<br /> |Abstract=&lt;p&gt;We present a neural network based approach for identifying salient features for classification in neural network committees. Our approach involves neural network training with an augmented cross-entropy error function. The augmented error function forces the neural network to keep low derivatives of the transfer functions of neurons of the network when learning a classification task. Feature selection is based on two criteria, namely the reaction of the cross-validation data set classification error due to the removal of the individual features and the diversity of neural networks comprising the committee. The algorithm developed removed a large number of features from the original data sets without reducing the classification accuracy of the committees. By contrast, the accuracy of the committees utilizing the reduced feature sets was higher than those exploiting all the original features. © Springer-Verlag Berlin Heidelberg 2003.&lt;/p&gt;<br /> |Opponents=<br /> |Supervisors=<br /> |Examiners=<br /> |Patent=<br /> |ThesisLevel=<br /> |Credits=<br /> |Programme=<br /> |Subject=<br /> |Uppsok=<br /> |DefencePlace=<br /> |DefenceLanguage=<br /> |DefenceDate=<br /> |CreatedDate=2018-04-06<br /> |PublicationDate=2018-04-06<br /> |LastUpdated=2018-04-06<br /> |diva=http://hh.diva-portal.org/smash/record.jsf?searchId=1&amp;pid=diva2:1195746}}<br /> &lt;div style=&#039;display: none&#039;&gt;<br /> <br /> == Keep all hand-made modifications below ==<br /> &lt;/div&gt;{{PublicationDisplayTemplate}}</div>

Slawek