Numerous industries require accurate weather forecasts; robust algorithms, methodologies, and procedures are required. To determine the most effective algorithm for weather forecasting, it is crucial to compare many algorithms. This study aims to identify the most effective machine learning algorithm for weather forecasting by comparing the efficiency of numerous prominent algorithms. A database is compiled and analyzed for this comparative study. Various machine learning algorithms on the dataset are evaluated to determine which generates the most accurate forecasting results. The nine techniques employed are Cataboost, Gradient Boosting, XGBoost, GaussianNB, LGBM, logistic regression, SVM, Adaboost, and KNN. The dataset has been divided into training and testing sets to train and evaluate algorithms effectively. The most accurate weather prediction algorithm can be identified by comparing the results of multiple algorithms. According to the results, Catboost has the highest accuracy, at 70.12%. In addition, the study’s original purpose has been expanded to include recommendations for the finest crops to produce in each climate. Through comparative analysis, the algorithm with the highest degree of precision may be taught as the most effective weather prediction algorithm for machine learning. Overall, the comparative analysis of machine learning algorithms for weather forecasting has substantially contributed by highlighting the merits and limitations of various approaches. Consequently, academicians and practitioners can significantly improve their decision-making process with access to comprehensive weather prediction algorithms. As a result, industries reliant on weather data can benefit from more precise forecasts, leading to enhanced outcomes across a diverse range of sectors.