Аннотация.

Рассматривается задача исследования системы управления заданного множества объектов, каждый из которых характеризуется фиксированным (исходным) набором разнородных параметров. Для этого исследуется структура расположения управляемых объектов в пространстве информативных параметров. Для выявления такой структуры разработан комплекс алгоритмов интеллектуального анализа данных (ИАД), а также процедур экспертной коррекции. Проведен теоретический анализ алгоритмов ИАД, доказаны теоремы об их сходимости.

Ключевые слова:

интеллектуальный структурно-классификационный анализ данных, информативные параметры, начального разбиение, выбор числа классов, заполнение пропущенных наблюдений, процедуры экспертной коррекции.

Стр. 36-46.

Полная версия статьи в формате pdf. 

J.A. Dorofeyuk, A.A. Dorofeyuk, Pokrovskaya I.V., A.G. Spiro

"Data mining methods in the study of complex control systems"

Abstract. The control complex system study problem is considered for a given set of objects, each of which is characterized by the fixed (initial) set of heterogeneous parameters. In the paper, to solve this problems, it is proposed to investigate the structure relationship of these objects in the informative parameters space. This allows to significantly increase the analysis efficiency of the system functioning and the stability of the procedures for making management decisions. To identify such patterns special mining complicated data algorithms complex and expert correction procedures were designed. The complex includes the following algorithms: structural-classification data analysis (SCDA), selection of informative parameters, select the initial conditions for classification algorithms, select the «optimal» number of classes, completing missing observations, as well as procedures of expert correction of the results for these algorithms. The theoretical analysis of different variants SCDA algorithm was carry out, the algorithm convergence theorems to the local extremum of the appropriate quality criterion were proved.

Keywords: intellectual structural-classification data analysis, informative parameters, initial partitioning, the number of classes selection, filling in missing observations expert correction procedures.

References

1. Dorofeyuk, A. A., Baumann E. V., Dorofeyuk J. A. Intelligent information processing methods based on the stochastic approximation algorithms. // Mathematical methods of pattern recognition. 15th international conference: Collection of reports. M.: MAKS PRESS, 2011. С. 108-112. (in Russian)
2. M. D. Goldovskaya, J. A., Dorofeyuk, N. E.Kiseleva Structural analysis methods in applications of the time series study. // Control Sciences, 2013, No 3. pp. 33-41. (in Russian)
3. Braverman E. M., Muchnik I. B. Structural methods of empirical data processing. Moscow: Nauka, 1983. (in Russian)
4. Dorofeyuk A. A. The expert-classification analysis methodology in the complex data control and processing problems (history and prospects). // Control Sciences, 2009. No 3.1. pp. 19-28. (in Russian)
5. Dorofeyuk A. A., Pokrovskaya I. V., Chernyavsky, A. L., Expert analysis and improvement control systems methods // Automation and remote control. 2004, No10. – pp. 172 – 188. (in Russian)
6. Dorofeyuk, A. A., Dorofeyuk J. A., Pokrovskaya I. V., Chernyavsky A. L. Independent multiple-choice expertise method and its use in the applied problems solving. / Managing the large-scale systems development (MLSD’2013): Proceedings of the Seventh international conference. Vol. 1. – M.: ICS RAS, 2013.  – pp. 260-271. (in Russian)
7. Крамер Г. Математические методы статистики. / Пер. с англ. А.С.Монина и А.А.Петрова под редакцией академика А. Н. Колмогорова./ ? М.: «МИР» Kramer G. Mathematical methods of statistics. / Trans. from English. by A. S. Monin and A. A. Petrov, edited by academician A. N. Kolmogorov./ ? M.: "MIR", 1975. (in Russian)
8. Dorofeyuk, J. A., Goldovskaya M. D., Spiro A. G. The computer implementation and simulation features of algorithms for complex data mining. // Managing the large-scale systems development (MLSD’2013): Proceedings of the Seventh international conference. Vol. 2. – M.: ICS RAS, 2013. – pp. 328-331. (in Russian)
9. Bellman R. Dynamic Programming and Lagrange Multipliers. / Proc. Nat. Acad. of Sc. USA, 1956, vol.42, pp.767-769.