ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 12. Vol. 24. 2018

DOI: 10.17587/it.24.799-804

V. D. Korepanov, Bachelor's Degree student, e-mail: korepanov.viacheslav@yandex.ru, V. P. Kulagin, Professor, e-mail: Kulagin_VP@mgupi.ru, R. F. Halabiya, Associate Professor, e-mail: rustam-capitan@mail.ru, MIREA — Russian Technological University

The Use of Neural Networks in the Modification of the Genetic Algorithm

Genetic algorithms are increasingly used in the development of modern software, in artificial intelligence systems, in the solution of optimization problems, in the construction of artificial neural networks, as well as in other branches of knowledge. An actual and promising direction is the joint use of genetic algorithms and neural networks. As a result hybrid systems are formed to solve specific practical problems. In the article the approach to modification of the genetic algorithm by elements of neural networks is considered. The purpose of this modification is to construct a new more efficient algorithm that surpasses the classical genetic algorithm in a certain class of problems. Efficiency analysis of the proposed algorithm is based on the results obtained in the "electronic nose" investigation (devices for odors recognition of a wide chemicals class).
Keywords: genetic algorithm, individual, chromosome, gene, reproduction, mutation, fitness, neural network

P. 799–804

To the contents