TITLE:
Using Genetic Algorithms for Solving the Comparison-Based Identification Problem of Multifactor Estimation Model
AUTHORS:
Andraws Swidan, Shmatkov Sergey, Bulavin Dmitry
KEYWORDS:
Genetic Algorithm; Comparatory Identification; Fitness-Function; Chromosome; Crossover; Mutation
JOURNAL NAME:
Journal of Software Engineering and Applications,
Vol.6 No.7,
June
28,
2013
ABSTRACT:
In this paper the statement and the
methods for solving the comparison-based structure-parametric identification
problem of
multifactor estimation model are addressed. A new method that combines
heuristics methods with genetic algorithms is proposed to solve the problem. In order to overcome some
disadvantages of using the classical utility functions, the use of nonlinear
Kolmogorov-Gabor polynomial, which contains in its composition the first as
well as higher characteristics
degrees and all their possible combinations is proposed in this paper. The use
of nonlinear methods for identification of the multifactor estimation model showed that the use of
this new technique, using as a utility function the nonlinear Kolmogorov-Gabor
polynomial and the use of genetic algorithms to calculate the weights, gives a
considerable saving in
time and accuracy performance. This method is also simpler and more evident for
the decision maker (DM) than other methods.