Hybrid Approach Combining Model-Based Method with the Technology of Machine Learning for Forecasting of Dangerous Weather Phenomena

摘要

The paper is a continuation of the works [1-4] where has been shown how the technologies of machine learning and online analytical processing (OLAP) could be used in conjunction with the numerical model of convective cloud for forecasting dangerous convective phenomena such as thunderstorm, heavy rainfall and hail. We study specifically the possibility of making predictions via a hybrid approach that combines the predictive numerical model of convective cloud with the modern methods of big data processing. We overview the existing examples of using of machine learning tools for weather forecasting and discuss the range of their applicability.