English
Język Polski
ESTIMATING THE POSSIBILITY OF A FIRE USING SATELLITE MISSIONS
Abstract
Space has been in the area of human interest for decades. Due to the multitude of
satellite missions, we receive huge amounts of data every day that are used in many
ways. The advantages of this data collection method include, among others, free access
and the ability to record data without the need for physical sensors placed on
Earth. This paper aims to present an attempt to use satellite missions to estimate the
possibility of a fire in a specific area based on many environmental factors and the
use of machine learning. The data used in this publication are based on satellite missions
operating for the Copernicus program. The study area covers the territory of
the Republic of Poland. The result of several models and Machine Learning methods
were compared, including: under-sampling, over-sampling, catboost, randomforest.
The research developed can serve to further develop work on a system that predicts
the occurrence of a fire in an area. Through this research, it is possible to identify the
most important relationships and features that correlate with the occurrence of fire.
Keywords:
Copernicus, possibility of a fire, satellites, machine learningDetails
References
Statistics
Authors
Citation rules
Licence
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
