Increasing the share of correct clustering of characteristic signal with random losses in self-organizing maps
Journal Title: Восточно-Европейский журнал передовых технологий - Year 2019, Vol 2, Issue 4
Abstract
<p class="Standard"><span lang="EN-US">Analysis of methods for optimizing algorithms of functioning of the Kohonen neural networks, self-organizing maps (SOM), in terms of training speed and percentage of correct clustering was made. Effective optimization of self-organizing maps was determined by the second criterion, the enhanced self-organizing incremental neural network (ESOINN). It was established that in the case of incomplete input signal, that is the signal with losses at unknown time points, the share of correct clustering is unacceptably low with any SOM algorithms, both basic and optimized.</span></p><p class="Standard"><span lang="EN-US">The incomplete signal was represented as the input vector of the neural network, the values of which are represented by a single array, that is without taking into consideration conformity of the moments of losses to the current values and without the possibility of determining these moments. A method for determining conformance of the incomplete input vector to the input layer of neurons to increase percentage of correct recognition was programmed and proposed. The method is based on finding the minimum distance between the current input vector and the vector of weights of each neuron. To reduce operating time of the algorithm, it was proposed to operate not with individual values of the input signal but their indivisible parts and the corresponding groups of input neurons. The proposed method was implemented for the SOM and ESOINN. To prove effectiveness of implementation of the basic algorithm of the SOM, its comparison with existing counterparts of other developers was made.</span></p><p class="Standard"><span lang="EN-US">A mathematical model was developed for formation of examples of complete signals of a training sample on the basis of reference curves of the second order and a training sample was generated. In accordance with the training sample, training of all neural networks implemented with and without the proposed method was made. A diagram of simulation of losses was developed and test samples were generated for computational experiments with incomplete signals.</span></p><span lang="EN-US">On the basis of experiments, efficiency of the proposed method for classification in terms of incomplete input signal on the basis of self-organizing maps was proved both for implementations of the basic algorithm of SOM and ESOINN</span>
Authors and Affiliations
Svitlana Shapovalova, Yurii Moskalenko
Keywords
Related Articles
- EP ID EP666296
- DOI 10.15587/1729-4061.2019.160670
- Views 40
- Downloads 0
An impact of the ladle lining on the refining of reinforced steel when blowing with powders
<p>When refining steel by blowing powders, the ladle lining material plays a serious role. The dependency graphs of the degree of desulfurization on the consumption of silicocalcium powder during blowing in ladles lined...
Analysis of possibilities for improving energy indicators of induction electric motors for propulsion complexes of autonomous floating vehicles
<p>This study proposes additional possibilities to improve, at minimal efforts, a number of technical-economic indicators of three-phase standard industrial induction motors (IM). We suggest that such engines should be u...
Studying the effect of formulation components on the physical-chemical properties of sweet sauces that contain physically-modified starches
<p>We have defined reference indicators for the consistency of sweet sauces depending on the content of starches, which has become a prerequisite for the development of a formulation for sweet sauces using physically-mod...
Development of a technology for the surface strengthening of barrel channels in the large-caliber artillery guns
<p>High temperatures and exposure to the chemical effect of powder gases that accompany artillery salvos actively destroy the surface layer of a barrel channel material. This disrupts the geometry of a barrel channel, th...
Design of fire-resistant heat- and soundproofing wood wool panels
<p>The conducted research revealed the possibility to manufacture heat- and soundproofing materials for the arrangement of buildings. Wood fibers, which are produced in the form of flat panels, are the raw materials for...