Informatics and Applications

2020, Volume 14, Issue 3, pp 109-118

OPTIMIZATION MODELS EXTRACTION FROM DATA

V. I. Donskoy

Abstract

The basic principles, methods and algorithms representing a new information technology for building optimization mathematical models from data (BOMD) are presented. This technology allows one to automatically build mathematical models of planning and control on the basis of use of precedents (observations) over objects that gives the chance to solve the problems of intellectual control and to define expedient behavior of economic and other objects in difficult environments. The BOMD technology allows one to obtain objective control models that reflect real-life relationships, goals, constraints, and processes. This is its main advantage over the traditional, subjective approach to control. Linear and nonlinear algorithms for synthesis of models based on precedent information are developed.

[+] References (15)

Mazurov, Vl. D. 1971. Primenenie metodov teorii raspo- znavaniya obrazov v optimal'nom planirovanii i upravlenii [Application of pattern recognition methods in optimal planning and management]. Tr. I Vsesoyuzn. konf. po optimal'nomu planirovaniyu i upravleniyu narodnym khozyaystvom [1st All-Union Conference on Optimal Planning and Management of the National Economy Proceedings]. Moscow. 1-49.
Zhuravlev, Yu. I. 1976. Ekstremal'nye algoritmy v matematicheskikh modelyakh dlya zadach raspoznavaniya i klassifikatsii [Extremal algorithms in mathematical models for discrimination and classification problems]. Dokl. Akad. NaukSSSR 231(3):532-535.
Eremin, 1.1., and V. D. Mazurov. 1979. Nestatsionarnye protsessy matematicheskogo programmirovaniya [Non- stationary processes of mathematical programming]. Moscow: Nauka. 288 p.
Rudakov, K. V. 1989. Ob algebraicheskoy teorii universal'nykh i lokal'nykh ogranicheniy dlya zadach klassifikatsii [On the algebraic theory of universal and local constraints for classification problems]. Raspoznavanie, klassifikatsiya, prognoz [Recognition, classification, forecast]. Moscow: Nauka. 176-201.
Vorontsov, K. V. 2014. Additive regularization for topic models of text collections. Dokl. Math. 89(3):301-304.
Makhina, G. A. 2000. Neyrosetevoy podkhod k zadacham slaboopredelennoy optimizatsii [Neural network approach to weakly defined optimization problems]. Iskusstvennyy intellekt [Artificial Intelligence] 2:145-148.
Taratynova, N.Yu. 2006. Postroenie optimizatsionnoy modeli po pretsedentnoy nachal'noy informatsii kak zadacha nelineynoy regressii [Optimization model con- structionby precedents as a nonlinear regression problem]. Iskusstvennyy intellekt [Artificial Intelligence] 2:238-241.
Donskoy, V. I. 2012. Sintez soglasovannykh lineynykh optimizatsionnykh modeley po pretsedentnoy informatsii: podkhod na osnove Kolmogorovskoy slozhnosti [Synthesis of coordinated linear optimization models according to precedential information: An approach based on Kolmogorov complexity]. Tavricheskiy vestnik informatiki i matematiki [Taurida J. Computer Science Theory Mathematics] 2:13-25.
Donskoy, V. I. 2017. Izvlechenie optimizatsionnykh modeley iz dannykh: podkhod na osnove reshayushchikh derev'ev i lesov [Extracting optimization models from data: An approach based on decision trees and forests]. Tavricheskiy vestnik informatiki i matematiki [Taurida J. Com-puter Science Theory Mathematics] 4:59-86.
Donskoy, V. 1.2018. Asynthesisof pseudo-boolean empirical models by precedential information. Vestnik Yuzhno- Ural'skogo gosudarstvennogo universiteta. Ser. Matemati- cheskoe modelirovanie i programmirovanie [Bull. South Ural State University. Ser. Mathematical Modelling Programming Computer Software] 11(2):96-107.
Nilsson, N.J. 1965. Learning machines. New York, NY McGraw-Hill. 137 p.
Rokach, L., and B. Scholkopf. 2004. Atutorial on support vector regression. Stat. Comput. 14(4):199-222.
Hornik, K., M. Stinchcombe, and H. White. 1990. Uni-versal approximation of an unknown mapping and derivatives using multilayer feedforward networks. Neural Networks 3(1):551-560.
Haykin, S. 2008. Neural networks and learning machines. New York, NY: Prentice Hall. 906 p.
Breiman, L., J. H. Friedman, R. Olshen, and C. J. Stone. 1984. Classification and regression trees. New York, NY Chapman and Hall. 369 p.

[+] About this article