Науковий вісник НГУ

Synthesis fundamentals of classifiers for technical systems of patterns recognition with the use of human’s models of emotional processes

• 
• 

User Rating:  / 1

PoorBest 

Details

Category: Information technologies, systems analysis and administration

Last Updated on 04 June 2015

Published on 29 March 2015

Hits: 4884

Authors:

A.S. Meniailenko, Dr. Sci. (Tech.), Professor, Luhansk Taras Shevchenko National University, Vice Rector, Luhansk, Ukraine

O.I. Zakhozhay, Cand.Sci. (Tech.), Associate Professor, Donbass State Technical University, Senior Lecturer of the Electronic Systems Department, Alchevsk, Ukraine

Abstract:

Purpose. Development of the emotional processes application method, by analogy with human, in technical systems of patterns recognition. It gives possibility to enhance the reliability and reduce the time spent on the classification.

Methodology. We have analyzed the appropriateness of use of different emotional processes in design of classifiers of technical systems of patterns recognition. The proposed information model of human memory has been generalized for the technical intelligent systems; it was based on the Atkinson-Shifrin concept. We suggest using of emotional components in the classification algorithms for the combined systems of patterns recognition. We have determined the ways of further investigations concerning the improvement of the classifiers design methods with the use of emotional processes.

Findings. Comparative analysis of human-made intelligent systems with human intelligence showed the classification results discrepancy connected with the additional emotional aspects which are not considered. The proposed information model is a generalization of the Atkinson-Shifrin model for the technical recognition systems. To make the recognition systems similar to the cognitive human apparatus, we have proposed improvement of memory model through the introduction of characteristics describing the emotional processes. This allows ranking signs for their placement in short-term memory. We have proposed to determine quantitative estimates of emotional stats as a differential of the objective function for each individual information channel. We have found out that the use of combined systems of patterns recognition requires comparison of emotional processes on various information channels.

Originality. We have proposed the information model of human memory that is a generalization of the Atkinson-Shifrin model for the technical recognition systems with characteristics describing the emotional processes, and the method for quantitative estimation of the level of emotional processes in technical systems of patterns recognition. We have defined the concept of emotional processes use in combined systems of patterns recognition.

Practical value. The use of proposed solutions allows implementing rational aggregate of emotional components in the technical systems of patterns recognition. For defined conditions it allows us to increase the classification accuracy and reduce the time complexity of this process.

References:

1. Zhongzhe Xiao, Emmanuel Dellandrea, Weibei Dou and Liming Chen, (2011), “Classification of emotional speech based on an automatically elaborated hierarchical classifier”, International Scholarly Research Network ISRN Signal Processing, Article ID 753819, DOI: 10. 5402/2011/753819.

2. Ruzova, T.A. (2012), “Dispersion aggregated objects images skeletonization”, Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu, no. 1, pp. 107–112.

 Рузова Т.А. Построение скелетов изображений агрегированных объектов дисперсий / Т.А. Рузова // Научный вестник Национального горного университета. – 2012. – № 1. – С. 107–112.

3. Mazurenko, I.L. (1998), Kompiuternye sistemy raspoznavaniya rechi [Computer Systems of Speech Recognition], vol. 3, issue 1–2, Phasis, Moscow, Russia.

 Мазуренко И.Л. Компьютерные системы распознавания речи / Мазуренко И.Л. – М.: Фазис,1998. –Т. 3. – Вып. 1–2. – 483с.

4. Simankov, V.S. (1999), Adaptivnoye upravleniye slozhnymi sistemami na osnove teorii raspoznavaniya obrazov [Adaptive Management of Complex Systems Based on the Theory of Patterns Recognition], Monograph, Technical University of Kuban State Technological University, Krasnodar, Russia.

 Симанков В.С. Адаптивное управление сложными системами на основе теории распознавания образов: монография / В.С. Симанков, Е.В. Луценко – Краснодар: Техн. ун-т Кубан. гос. технол. ун-та, 1999. – 318 с.

5. A hlem Othmani, Lew F.C., Lew Yan Voon, Christophe Stolz and Alexandre Piboule, (2013), “Single tree species classification from Terrestrial Laser Scanning data for forest inventory”, Pattern Recognition Letters, Vol. 34, pp. 2144–2150.

6. Antoshchenko, N.I., Okalelov, V.N. and Pavlov, V.I. (2010), Geomekhanicheskiye protsessy i prognoz dinamiki gazovydeleniya pri vedenii ochistnykh rabot v ugolnykh shakhtakh [Geomechanical Processes and Prediction of Gassing Dynamics When Brushing in Coal Mines] Monograph, DonSTU, Alchevsk, Ukraine.

 Антощенко Н.И. Геомеханические процессы и прогноз динамики газовыделения при ведении очистных работ в угольных шахтах: монография / Антощенко Н.И., Окалелов В.Н., Павлов В.И. – Алчевск: Дон ГТУ, 2010. – 451 с.

7. Norman Poh, Arun Ross, Weifeng Lee, Josef Kittler, (2013), “A user-specific and selective multimodal biometric fusion strategy by ranking subjects”, Pattern Recognition Journal, Vol. 46, Issue 12, pp. 3341–3357.

8. Cohen, A. and Glicksohn, A. (2011), “The role of Gestalt grouping principles in visual statistical learning”, Attention, Perception & Psychophysics, vol. 73, pp. 708–713.

9. Ryabenky, V.M.and Zakhozhay, O.I. (2012), “Combined systems of patterns recognition”, Problemy Informatsiinykh Tekhnologiy, vol. 01(009), pp. 156–160.

 Рябенький В.М. Комбіновані системи розпізнавання образів / В.М. Рябенький, О.І. Захожай // Проблеми інформаційних технологій. – 2011. – № 01 (009). – С. 156–160.

10. Zakhozhay, O.I. (2013), “The rational aggregate selection of informative patterns in the combined recognition systems”, Elektrotekhnichni ta Kompiuterni Systemy, vol. 09(85), pp. 186–192.

 Захожай О.І. Селекція раціональної сукупності інформативних образів у комбінованих системах розпізнавання / О.І. Захожай // Електротехнічні та комп’ютерні системи. – 2013. – № 09(85). – С.186–192.

2015_1_meniailenko
2015-03-27  401.89 KB  963

• < Prev