Beschreibung
There are a lot of different prediction methods exist in our days. Many of them work well in a specific situation. But when the situation deals with the big volume of data and complicated classification analysis there is a good opportunity to use the machine learning for the prediction. Previously there were a lot of different ways of the machine learning applying in the ecological modeling and prediction. But the method of the ecosystem condition modeling by the geofields analysis wasn't applied for the artificial intelligence. Previously such kind of the classification was made only by the human expert. The described method automatizes expert analysis of this problem and adds a new kind of the application of the machine learning in the environmental science. In this work the method of the ecosystem stability level assessment using machine learning methods was developed. It combines ways of the data preprocessing for the solving of this problem and the using of concrete algorithms for the best modeling of the process. The final model significance was around 80% in different testing datasets.
Autorenporträt
Rukavitsyn Vadim, Master of Ecology: Studied ecology and geology in Russian Geological Prospecting University and artificial intelligence in Mendel University in Brno. Deputy Head of Environmental Assessment and Monitoring Department, Moscow.
Herstellerkennzeichnung:
OmniScriptum SRL
Str. Armeneasca 28/1, office 1
2012 Chisinau
MD
E-Mail: info@omniscriptum.com
