Forecasting forest fire burning area using machine training
https://doi.org/10.23947/2541-9129-2019-3-17-22
Abstract
The objective of this article is to create and train an artificial neural network based on a data set containing various climatic parameters and future fire area as an output parameter that the authors intend to predict. Such a “set” of data is usually available for research and study. Before training the neural network model, the data set is divided into two samples: a sample for training, which is about 90% of the set; and a sample for testing the trained model. In setting the task, the authors select and analyze the known data on the fires that occurred in Montesinho Park, compare the models trained on these data with and without normalization. As a result, two examples are given of a qualitative demonstration of graphs of absolute error changes of fire areas, which are projected using the created and trained model.
About the Authors
V. A. Filippenko
Don State Technical University
Russian Federation
Russian Federation
Filippenko Viktor Aleksandrovich, student
1, Gagarin sq., Rostov-on-Don, 344000, Russia
A. V. Zotov
Don State Technical University
Russian Federation
Russian Federation
Zotov Aleksey Vyacheslavovich, student
1, Gagarin sq., Rostov-on-Don, 344000, Russia
References
1. Cortez, P.A. Morais, A. Data Mining Approach to Predict Forest Fires Using Meteorological Data. Dep. Information Systems/Algoritmi R&D Centre University of Minho. Available at: http://www3.dsi.uminho.pt/pcortez/fires.pdf (accessed: 10.06.2019).
2. Biblioteki dlya glubokogo obucheniya: Keras. [Libraries for deep learning: Keras.] Open Data Science. Available at: https://habr.com /EN/company /ods/blog /325432/ (accessed: 10.06.2019) (in Russian).
3. Lutts, M. Izuchaem Python. [Learning Python.] St. Petersburg: Symbol-Plus, 2011, 1280 p. (in Russian).
4. Myuller, A., Gvido, S. Vvedenie v mashinnoe obuchenie s pomoshchyu Python. Rukovodstvo dlya spetsialistov po rabote s dannymi. [Introduction to machine learning using Python. A guide for professionals working with data.] St. Petersburg: OOO "Alfa-kniga", 2017, 480 p. (in Russian).
5. Gubenko, I.M., Rubinshteyn, K.G. Sravnitel’nyy analiz metodov rascheta indeksov pozharnoy opasnosti. [Comparative analysis of methods for calculating fire hazard indices.] Gidrometeorologicheskiy nauchno-issledovatel’skiy tsentr Rossiyskoy Federatsii. [Hydrometeorological Research Center of the Russian Federation.] Available at: http:// method.meteorf.ru /publ/tr/tr347 /gubenko.pdf (accessed: 10.06.2019) (in Russian).
6. Kin, B.A. Mashtabirovanie funktsiy s pomoshchyu scikit-learn. [Scaling functions using scikit-learn.] Available at: http:// benalexkeen.com/ feature-scaling-with-scikit-learn/ (accessed: 10.06.2019) (in Russian).
7. Esipov, Yu.V., Dzilyadzhi, M.S., Cheremisin, A.I. Logicheskoe i parametricheskoe modelirovanie predposylok i ustanovlenie mery opredelennosti realizatsii proisshestviya v siateme. [Logical and parametric modeling of perquisites and the establishment of measures of certainty of realization of the incident in the system.] Bezopasnost’ v tekhnosfere 2017, vol. 6, no. 2, pp. 3-11 (in Russian).
8. Esipov, Yu.V., Mishenkina, Yu.S., Cheremisin, A.I. Modeli i pokazateli tekhnosfernoy bezopasnosti. [Models and indicators of technosphere safety.] Moscow: INFRA, 2018, 154 p. (in Russian).
Review
For citations:
Filippenko V.A., Zotov A.V. Forecasting forest fire burning area using machine training. Safety of Technogenic and Natural Systems. 2019;(3):17-22. https://doi.org/10.23947/2541-9129-2019-3-17-22
Views:
249
Email this article 