Construction of models predicting through artificial neurons networks

О.Б. Іванець, О.В. Букрєєва, М.В. Дворнік

Abstract

 In this paper the possibility of building prediction models by using artificial neural networks using the most common software products designed to meet the challenges of this type. The analysis of the results. Display disadvantages and advantages of each software product to solve such problems.

Keywords

нейрон; штучні нейронні мережі; програмний продукт; діагностичні дані

References

Kevin Swingler Applying Neural Networks. A practical Guide – http://matlab.exponenta.ru/neuralnetwork/book4/3_2.php

Neocleous, CK Anastasopoulos, P. Nikolaides Neural networks to estimate the risk for preeclampsia occurrence // Neural Networks. – 2009. – P. 2221–2225.

Upadhyaya B. R. Application of neural networks for sensor validation and plant monitoring / B. R. Upadhyaya, E. Eryutek // Neural Technology. – No 97 – 1992. – Р. 170–176.

Frean M. The upstart algorithm: A method of constructing and training feedforward neural networks / M. Frean // Neural Computation. – No2. – 1990. – P. 198–209.

Cottrell G. W. Image compression by neural network. An example of extensional programming / G. W. Cottrell, P. Munroe. Technical report, University of California. San Diego, Institute of Cognitive Science.– 1987.

Full Text: PDF