Development of a Automated Environmental Monitoring System with Forecasting

All environmental and monitoring systems are most often based on the accumulated statistics of various parameters of temperature, harmful emissions, humidity, carbon dioxide. This paper describes methods of implementing machine learning algorithms for forecasting problems within a specifically developed ASMOS system in Russia. The proposed algorithm makes it possible to consider the change in temperature for several hours in advance, as well as days. Later this algorithm can be implemented in the system for assessing hazardous environmental situations, where a comparison of hazardous factors affecting the environment carried out.