hh.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Malmqvist, Kerstin
Publications (3 of 3) Show all publications
Verikas, A., Bacauskiene, M. & Malmqvist, K. (2003). Selecting salient features for classification committees. In: Kaynak, O Alpaydin, E Oja, E Xu, L (Ed.), Artificial Neural Networks and Neural Information Processing — ICANN/ICONIP 2003: . Paper presented at Joint International Conference on Artificial Neural Networks (ICANN)/International on Neural Information Processing (ICONIP), JUN 26-29, 2002, ISTANBUL, TURKEY (pp. 35-42). Heidelberg: Springer Berlin/Heidelberg, 2714
Open this publication in new window or tab >>Selecting salient features for classification committees
2003 (English)In: Artificial Neural Networks and Neural Information Processing — ICANN/ICONIP 2003 / [ed] Kaynak, O Alpaydin, E Oja, E Xu, L, Heidelberg: Springer Berlin/Heidelberg, 2003, Vol. 2714, p. 35-42Conference paper, Published paper (Refereed)
Abstract [en]

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.

Place, publisher, year, edition, pages
Heidelberg: Springer Berlin/Heidelberg, 2003
Series
Lecture Notes in Computer Science, ISSN 0302-9743 ; 2714
Keywords
Errors, Neural networks
National Category
Bioinformatics (Computational Biology) Telecommunications Bioinformatics and Systems Biology
Identifiers
urn:nbn:se:hh:diva-35784 (URN)10.1007/3-540-44989-2_5 (DOI)000185378100005 ()2-s2.0-21144434169 (Scopus ID)978-3-540-40408-8 (ISBN)978-3-540-44989-8 (ISBN)
Conference
Joint International Conference on Artificial Neural Networks (ICANN)/International on Neural Information Processing (ICONIP), JUN 26-29, 2002, ISTANBUL, TURKEY
Available from: 2018-04-06 Created: 2018-04-06 Last updated: 2018-04-06Bibliographically approved
Verikas, A., Lipnickas, A. & Malmqvist, K. (2002). Selecting neural networks for making a committee decision. In: Dorronsoro, J R (Ed.), ARTIFICIAL NEURAL NETWORKS - ICANN 2002: . Paper presented at 12th International Conference on Artifical Neural Networks (ICANN 2002), AUG 28-30, 2002, MADRID, SPAIN (pp. 420-425). Berlin: Springer Berlin/Heidelberg, 2415
Open this publication in new window or tab >>Selecting neural networks for making a committee decision
2002 (English)In: ARTIFICIAL NEURAL NETWORKS - ICANN 2002 / [ed] Dorronsoro, J R, Berlin: Springer Berlin/Heidelberg, 2002, Vol. 2415, p. 420-425Conference paper, Published paper (Refereed)
Abstract [en]

To improve recognition results, decisions of multiple neural networks can be aggregated into a committee decision. In contrast to the ordinary approach of utilizing all neural networks available to make a committee decision, we propose creating adaptive committees, which are specific for each input data point. A prediction network is used to identify classification neural networks to be fused for making a committee decision about a given input data point. The jth output value of the prediction network expresses the expectation level that the jth classification neural network will make a correct decision about the class label of a given input data point. The effectiveness of the approach is demonstrated on two artificial and three real data sets.

Place, publisher, year, edition, pages
Berlin: Springer Berlin/Heidelberg, 2002
Series
Lecture Notes in Computer Science, ISSN 0302-9743 ; 2415
Keywords
Neural networks
National Category
Bioinformatics (Computational Biology) Telecommunications Bioinformatics and Systems Biology
Identifiers
urn:nbn:se:hh:diva-35788 (URN)10.1007/3-540-46084-5_68 (DOI)000181441900068 ()2-s2.0-33745928507 (Scopus ID)978-3-540-44074-1 (ISBN)978-3-540-46084-8 (ISBN)
Conference
12th International Conference on Artifical Neural Networks (ICANN 2002), AUG 28-30, 2002, MADRID, SPAIN
Available from: 2018-04-05 Created: 2018-04-05 Last updated: 2018-04-05Bibliographically approved
Verikas, A., Malmqvist, K., Bacauskiene, M. & Bergman, L. (2000). Monitoring the de-inking process through neural network-based colour image analysis. Neural computing & applications (Print), 9(2), 142-151
Open this publication in new window or tab >>Monitoring the de-inking process through neural network-based colour image analysis
2000 (English)In: Neural computing & applications (Print), ISSN 0941-0643, E-ISSN 1433-3058, Vol. 9, no 2, p. 142-151Article in journal (Refereed) Published
Abstract [en]

This paper presents an approach to determining the colours of specks in an image of a pulp being recycled. The task is solved through colour classification by an artificial neural network. The network is trained using fuzzy possibilistic target values. The number of colour classes found in the images is determined through the self-organising process in the two-dimensional self-organising map. The experiments performed have shown that the colour classification results correspond well with human perception of the colours of the specks.

Place, publisher, year, edition, pages
New York, USA: Springer-Verlag New York, 2000
Keywords
Classification, Colour image processing, Fuzzy sets, Neural networks, Self-organising map, Classifier networks, Graphic arts, Fuzzy, Segmentation
National Category
Engineering and Technology
Identifiers
urn:nbn:se:hh:diva-3544 (URN)10.1007/s005210070025 (DOI)000088547900009 ()2-s2.0-0034336492 (Scopus ID)
Available from: 2010-01-13 Created: 2009-12-01 Last updated: 2018-03-23Bibliographically approved
Organisations

Search in DiVA

Show all publications