Time series forecasting based on machine learning methods for ensuring natural and technosphere safety

Babenyshev S.V., Ph.D. of Physical and Mathematical Sciences; Malyutin O.S.; Materov E.N., Ph.D. of Physical and Mathematical Sciences
FSBEE HE Siberian Fire and Rescue Academy EMERCOM of Russia

The main purpose of the article is to give a review of some capabilities of time series analysis and modeling using examples of predicting the number of fires and the level of river flooding using modern machine learning methods in the R programming language environment. The peculiarity of this simulation is the possibility of using several models at the same time, which allows to automatically select models with the lowest bias errors.

Открыть » // Ссылка для цитирования этой статьи

Ссылка для цитирования статьиЗакрыть
link…

Скопировать в буфер