Design of the universal classifier as a RBF network based on the CART solution tree
DOI:
https://doi.org/10.15587/1729-4061.2017.108976Keywords:
universal classifier, neural network, RBF network, CART Solution Tree, decision-making supportAbstract
The aim of the paper was to develop a universal classifier in the form of a radial basis function network (RBF network) based on the Gaussian function and the CART Solution Tree. The examples of diseases diagnostics classifier were considered. It is noted that during the classifier development, it is necessary to determine the number of RBF neurons and the values of parameters of these neurons (centre, dispersion). For this purpose, a method that allows splitting the space of features into relatively homogeneous domains in the form of hyperparallelepipeds, each of which is associated with one of the RBF neurons, is proposed. The number of RBF neurons and parameters of these neurons are determined automatically directly based on the CART Solution Tree.
As a result of the research, it was found that the proposed classifiers show the highest efficiency on the learning set with a minimal Solution Tree reduction (accuracy from 80 % to 95 %). It was shown that for two and more classes the accuracy of these classifiers on the test set makes 79 % and more, however, provided that the appropriate data sample for the learning set is selected. The possibility of using the RBF network based on the Gaussian function and the CART Solution Tree in the healthcare system for the diseases diagnostics and medical systems (or devices) assistance during decision-making support was proved.
The obtained results could be further applied to improve the universal classifier development method based on the RBF networkReferences
- Dobrovska, L. M., Dobrovska, I. A. (2015). Teoriya ta praktyka neironnykh merezh. Kyiv: NTUU «KPI» Vyd-vo «Politekhnika», 396.
- Osovskiy, S. (2004). Neyronnye seti dlya obrabotki informaciy. Moscow: Finansy i statistika, 343.
- Haykin, S. (2006). Neyronnye seti. Moscow: Izdatel'skiy dom «Vil'yams», 1104.
- Hagan, M., Demuth, H., Beale, M. (2010). Neural Network Design. USA: Colorado University Bookstore.
- Kitayama, Y., Kitayama, S., Yamazaki, K. (2008). Pattern Classification by RBF Network. The Proceedings of Design & Systems Conference, 2008.18, 330–332. doi: 10.1299/jsmedsd.2008.18.330
- Kubat, M. (1998). Decision trees can initialize radial-basis function networks. IEEE Transactions on Neural Networks, 9 (5), 813–821. doi: 10.1109/72.712154
- Safish, M., Joseph, R. (2012). Data classification with neural classifier using radial basis function with data reduction using hierarchical clastering. ICTACT Journal on Soft Computing, 2 (3), 348–352. doi: 10.21917/ijsc.2012.0054
- Jia, W., Zhao, D., Ding, L. (2016). An optimized RBF neural network algorithm based on partial least squares and genetic algorithm for classification of small sample. Applied Soft Computing, 48, 373–384. doi: 10.1016/j.asoc.2016.07.037
- Jang, J.-S., Sun, C.-T., Mizutani, E. (1997). Neuro-fuzzy and soft computing: a computation approach to learning and machine intelligence. USA, 614.
- Blake, C. L., Merz, C. J. (1998). UCI Repository of machine learning databases. University of California Irvine, Department of Information and Computer Science, Available at http://archive.ics.uci.edu/ml/datasets.html
- Faynzil'berg, L. (2009). Garantirovannaya ocenka ehffektivnosti diagnosticheskih testov na osnove usilennogo ROC-analiza. Upravlyayushchie sistemy i mashiny, 5, 3–13.
- Flach, P., Wu, S. (2003). Repairing concavities in ROC curves. Proc. UK Workshop on Comp. Intel, 38–44.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2017 Lyudmila Dobrovska, Iryna Dobrovska
This work is licensed under a Creative Commons Attribution 4.0 International License.
The consolidation and conditions for the transfer of copyright (identification of authorship) is carried out in the License Agreement. In particular, the authors reserve the right to the authorship of their manuscript and transfer the first publication of this work to the journal under the terms of the Creative Commons CC BY license. At the same time, they have the right to conclude on their own additional agreements concerning the non-exclusive distribution of the work in the form in which it was published by this journal, but provided that the link to the first publication of the article in this journal is preserved.
A license agreement is a document in which the author warrants that he/she owns all copyright for the work (manuscript, article, etc.).
The authors, signing the License Agreement with TECHNOLOGY CENTER PC, have all rights to the further use of their work, provided that they link to our edition in which the work was published.
According to the terms of the License Agreement, the Publisher TECHNOLOGY CENTER PC does not take away your copyrights and receives permission from the authors to use and dissemination of the publication through the world's scientific resources (own electronic resources, scientometric databases, repositories, libraries, etc.).
In the absence of a signed License Agreement or in the absence of this agreement of identifiers allowing to identify the identity of the author, the editors have no right to work with the manuscript.
It is important to remember that there is another type of agreement between authors and publishers – when copyright is transferred from the authors to the publisher. In this case, the authors lose ownership of their work and may not use it in any way.
