Analisis Klasifikasi Curah Hujan Kabupaten Banyuwangi Menggunakan Machine Learning

Abstract

This study analyzes rainfall patterns in Banyuwangi Regency using machine learning classification methods to support hydrometeorological disaster mitigation efforts. Rainfall data from 2019 to 2023 were obtained from the Meteorology, Climatology, and Geophysics Agency (BMKG) and used to classify rainfall into four categories: low, moderate, high, and extreme. The study applied several preprocessing steps including feature elimination, missing value handling, one-hot encoding for categorical features, SMOTE for class balancing, and feature normalization. Three machine learning algorithms were employed: Decision Tree (DT), Support Vector Machine (SVM), and Random Forest (RF). Model performance was evaluated using accuracy, precision, and recall metrics with various data splits and 10-fold cross-validation. The results show that the Random Forest algorithm achieved the highest performance with an average accuracy of 97%, outperforming Decision Tree (94%) and Support Vector Machine (86%). The findings indicate that ensemble models such as Random Forest are more robust and reliable for rainfall classification tasks. This research provides a foundation for early warning systems and can be expanded using additional environmental data and more advanced classification techniques in the future.