Амиргалиев Е.Н., Искаков С.Х., Кучин Я.В., Мухамедиев Р.И., Уалиева И.М., Мухамедиева Е.Л. Распознавание пород на урановых месторождениях с использованием методов машинного обучения. Совместный выпуск: Вестник восточно-Казахстанского государственного университет/Вычислительные технологии.Институт вычислительных технологий Сибирского отделения РАН. Том «Информационные и телекоммуникационные технологии».- Усть-Каменогорск, ВКГТУ им. Д. Серикбаева, ИВТ, 2013. ISSN 1561-4212, 1560-7534.  С.232-240.
Uranium mining on Kazakhstan’s deposits is carried out by the method of subsurface well leaching-out. The economic indicators of the mining process depend on the speed and accuracy of geophysical data interpretation. The learning systems can be used for data interpretation, for example articial neural networks (ANN). The results of preliminary research have shown that using only ANN enables to reach in average 55% of degree of coincidence of interpreted data in comparison with expert estimations. Therefore to generate training samples and improve the quality of interpretation it’s required to develop scientic methods and algorithms. The research results on the application of ANN in problems of recognition of lithological types are described in this work.
Comparison metric and statistical classication algorithms with ANN are shown. The problem of building a classier using a variety of classication algorithms is dened. Keywords: intellectual systems, Geophysical research of boreholes, machine learning, articial neural network, uranium deposit, pre-processing data, learning sample, algorithms.