Cybernetics and Computer Engineering, 2022, 3(209)
International Research and Training Center for Information Technologies
and Systems of the National Academy of Sciences of Ukraine
and Ministry of Education and Science of Ukraine.
40, Akad. Hlushkov av., Kyiv, 03187, Ukraine
PROBLEM OF CONSTRUCTING AN ONTOLOGICAL METAMODEL OF ITERATIVE ITIRATIVE GROUP METHOD OF DATA HANDLING ALGORITHMS
Introduction. Data volumes are permanently increasing and some new approaches are needed for storage and processing them considering the development and improvement of modern computers. This puts forward new requirements to automatic data processing tools and intelligent systems for analyzing information with taking into account its semantics.
The advantage of iterative GMDH algorithms is that they are able to work with a large number of arguments. The generalized iterative GMDH algorithm includes various former modifications of these algorithms. For example, algorithms of multilayer and relaxation types as well as varieties of iterative-combinatorial (hybrid) algorithms are diverse particular cases of the generalized one.
Metamodeling is the construction of generalized models of a certain group of objects (software tools, mathematical models, information systems). An ontological metamodel of the iterative GMDH algorithms was built using the Protege tools in order to structure knowledge in this subject area.
The purpose of the paper is to analyze the developed iterative GMDH algorithms and propose an approach to structuring knowledge оn iterative GMDH algorithms by building an ontological metamodel of this subject area.
Results. A retrospective analysis of the developed iterative GMDH algorithms іs carried out in the paper, their advantages and disadvantages are indicated. It is shown that the generalized iterative algorithm, whose special cases are both known and new varieties of multilayer, relaxation and iterative-combinatorial GMDH algorithms, makes it possible to compare the effectiveness of various algorithms and solve real modeling problems. Based on the results of this study, an ontological metamodel of iterative GMDH algorithms has been developed.
Conclusions. The advantage of iterative GMDH algorithms is that they allow processing big data sets. The generalized iterative algorithm allows forming typical architectures of previously developed modifications of these algorithms when setting up various operating modes of this algorithm. The construction of an ontological metamodel based on this one allows structuring knowledge on the available iterative algorithms making it possible to automate the design and use of specialized software tools for specific applied tasks.
Keywords: inductive modeling, GMDH, iterative algorithms, mathematical model, metamodeling, subject area, ontology
1 Ivakhnenko A.G., Stepashko V.S. Noise-immunity of modeling. Kiev: Naukova dumka, 1985. 216 p. (In Russian).
2 Ivakhnenko, A.G. Group method of data handling as competitor for the method of stochastic approximation. Soviet Automatic Control, 1968, no 3, pp. 58-72 (In Russian).
3 Spravochnik po tipovym programmam modelirovaniya / Red. Ivakhnenko A.G. Kiev: Tekhnika, 1980. 184 p. (In Russian).
4 Chandrasekaran B., Josephson J.R., Benjamins, R.V., Ontologies. What are ontologies, and why do we need them?” IEEE Intelligent Systems and their Applications. 1999, V. 14. Iss. 1. pp. 20-26.
5 Stepashko V.S., Bulgakova A.S., “http://usim.org.ua/arch/2013/2/3.pdf” The Generalized Iterative Algorithm of the Group Method of Data Handling. Upravlyayushchie Sistemy i Mashiny, 2013, no 2, pp. 5-17 (In Russian).
6 Stepashko V., Bulgakova O., Zosimov V. Construction and Research of the Generalized Iterative GMDH Algorithm with Active Neurons. In: Advances in Intelligent Systems and Computing II. AISC book series, Volume 689. Berlin: Springer, 2017, pp. 474-491.
7 Ruy F.B., Guizzardi G., Falbo, R.A., Reginato, C.C., Santos, V.A. From reference ontologies to ontology patterns and back. Data & Knowledge Engineering, 2017, 109, pp. 41-69.
8 Savchenko Ye., Stepashko V. Metamodeling and metalearning approaches in inductive modeling tools. Preprint, [online]. Available at: https://easychair.org/publications/ preprint/6L1W [Accessed 23 Apr. 2018].
9 Flach P. Machine learning: the art and science of algorithms that make sense of data, Cambridge University Press, 2012. 396 p.
10 Savchenko Ye.A, Stepashko V.S.,. “Analysis of approaches to metalearning and metamodeling”. Inductive modeling of complex systems, Coll. sciences works. Kyiv: IRTCITS, 2017, Iss. 9, pp. 86-94 (In Ukrainian).
11 Pidnebesna H.A. Conceptual development of ontology for the design of inductive modeling. Inductive modeling of complex systems. Coll. sciences works. Kyiv: IRTCITS, 2013, 5, pp. 248-256 (In Ukrainian).
12 Valkman Yu.R. Ontologies: formal and informal. Report at the seminar “Pattern computer”. 08.11.2011, [online]. Available at: http://www.irtc.org.ua/image/seminars/archive/int [Accessed 18 Dec. 2017] (In Russian).
13 Stepashko V., Bulgakova O., Zosimov V. Construction and Research of the Generalized Iterative GMDH Algorithm with Active Neurons. In: Advances in Intelligent Systems and Computing II. AISC book series, 2017, V. 689, Berlin: Springer, pp. 474-491.
14 Pavlov A.V., Kondrashova N.V. On the Convergence of the Generalized Relaxation Iterative Algorithm for the Method of Group Consideration of Arguments. Upravlyayushchie Sistemy i Mashiny, 2012, no 3 (239), pp. 24-29, 38 (In Russian).
15 Pavlov A.V. “Generalized Relaxation Iterative GMDH Algorithm”. Inductive modeling of complex systems, Coll. sciences works. Kyiv: IRTCITS, 2011, Iss. 4, pp. 121-134 (In Ukrainian).
16 Ivakhnenko N.A., Marchev A.A. Self-organization of a mathematical model for long-term planning of construction and installation works. Automation. 1978, no. 3, pp. 12-18 (In Russian).
17 Sheludko O.I. GMDH algorithm with orthogonalized complete description for model synthesis based on the results of the planned experiment. 1974, no 5, pp. 32-42 (In Russian).
18 Yefimenko S., Stepashko V. Technologies of Numerical Investigation and Applying of Data-Based Modeling Methods”. Proceedings of the II International Conference on Inductive Modelling ICIM-2008, 15-19 September 2008, Kyiv, Ukraine. Kyiv: IRTCITS, pp. 236-240.
19 Yefimenko, S.M. Stepashko, V.S. Computer tests as an instrument for effectiveness investigation of modeling algorithms. Proceedings of International Workshop on Inductive Modelling (IWIM 2007), Prague: Czech Technical University, pp. 123-127.
20 Tyryshkin A.V., Andrakhanov A.A., Orlov A.A. GMDH-based Modified Polynomial Neural Network Algorithm”, Chapter 6 in Book GMDH-methodology and implementation in C (With CD-ROM). London: Imperial College Press, World Scientific, 2015, pp. 107-155.
21 Kordik P. Why Meta-learning is Crucial for Further Advances of Artificial Intelligence? [online]. Available at: [Accessed 18 Dec. 2020].
22 Stepashko V., Bulgakova O., Zosimov V.,. “Hybrid Algorithms for Self-Organizing Models for Predicting Complex Processes”. Inductive modeling of complex systems. Coll. sciences works. Kyiv: IRTCITS, 2010, pp. 236-246 (In Ukrainian).
23 Ivakhnenko A.G., Ivakhnenko G.A., Muller J.-A. Self-Organization of Neuronets with Active Neurons. Patt. Recognition and Image Analysis. 1994, 4 (4), pp. 177-188.
24 Bulgakova O.S., Stepashko V.S. Comparative Analysis of the Efficiency of Iterative GMDH Algorithms Using Computational Experiments. Visnyk CHDTU. 2011, no 1, pp. 41-44 (In Ukrainian).
HOW TO CITE:
Stepashko V.S., Savchenko-Syniakova Ye.А., Pidnebesna H.А. Problem of Constructing an Ontological Metamodel of Itirative Group Method of Data Handling Algorithms. Cybernetics and Computer Engineering, 2022, no 3(209), pp. 21-33.