Makine öğrenmesi teknikleri kullanarak güneş aktivitesi, kozmik ışınlar ve dünya iklimi arasındaki ilişkilerin analizi

The Earth's climate is part of a complicated system that can be affected by many different parameters, both internal and external. Important external forces on the climate are galactic cosmic rays (GCR) and the Sun. Some research has already been conducted to investigate the relationship between the climate and external forces such as the GCR and solar activity. However, the relations are quite complicated and buried into almost chaotic meteorological measurements. This thesis looks deeper into the interactions between them. The parameters used in the correlation analysis are GCR flux, Sunspot number (SSN), total solar irradiance (TSI), UV irradiance (UVI), and the Oceanic Niño Index (ONI) as the predictor variables; total cloud amount (TCA), low cloud amount (LCA), global mean temperature anomaly (GMTA), aerosol optical depth (AOD) and precipitation as the response variables. The analysis begins with standard statistical techniques and continues with multiple regression and machine learning methods for non-linear regression, such as random forests. Both geographical and temporal patterns have been investigated. This study shows that some parameters have a weak linear correlation, while a statistically significant non-linear relationship occurs between them. It can be concluded that the GCR-climate connection does exist, and these non-linear relations should be investigated further, specifically in certain regions of the World.