A Machine Learning-Based Model for Predicting Temperature Under the Effects of Climate Change

Regarding the climate changet "راجع, importance of forecasting weather conditions, especially temperaturest "راجع, is necessary to avoid climate change conditions and recommend precautions and instructions to deal with emergency climate changes. In this chapter, we utilized Machine Leaning (ML) tools to predict the temperature as a target related to 10 features includes Year, Month, CH₄, N₂O, CFC.12, CFC.11, Aerosols, TSI, MEI. Current research has shown that the average world temperature has increased over the last 100 years, according to the Climate Change dataset. The repercussions of sustained global temperature increase will be disastrous. Billions of people will be affected by increased sea levels and an increase in the frequency of extreme weather occurrences. In this chapter, we attempt to investigate the link between average world temperature and a number of other variables using ML regressors. We present Linear Regressiont "راجع (LR), Random Forest (RF) regressor, Decision Tree (DT) regressor, K-Nearest Neighbor (KNN) regressor, Support Vector Machinet "راجع (SVM) regressor, and Cat Boost Regressor (CBR) as ML regressors to predict the global temperature. The results indicated that the CBR achieved high results compared the recent ML approaches. The evaluation of the proposed model investigated that the Cat boost regressor achieved Mean Square Errort "راجع (MSE), Root Mean Square Error (RMSE), Minimum Absolute Error (MAE), the determination Coefficient R² are 0.003, 0.054, 0.0036, and 92.40%, respectively.