부산대학교 도서관

Rainfall prediction is vital for various applications such as agriculture, flood control, and water resource management. Accurate rainfall prediction can help farmers to make decisions about irrigation and crop selection. It can also help authorities to take necessary measures to prevent or mitigate the impact of floods. In the proposed study, multiple algorithms are examined and compared in terms of how accurately they predict rainfall. The accuracy of rainfall forecasts may be increased by using machine learning algorithms, which have shown considerable promise in predicting rainfall patterns. The authors solely used the AUC score and categorial standards to analyze six ML algorithms. Based on their accuracy score, AUC, precision, recall, and f1 score, authors compared algorithms including Random Forest, K-Nearest Neighbours (KNN), Logistic Regression, Cat Boost, Naive Bayes, and Gradient Boost. Since they provide a useful solution for categorization and prediction of the amount of rainfall as well as precise rate prediction, the applied algorithms are supervised learning algorithms.