Machine learning method with random forest model for flood hydrometeorological disaster mitigation in Demak

Sutopo, Joko and Pabbajah, Mustaqim and Juhansar and Rohmatika, Fiya Ainur and Abd Ghani, Mohd Khanapi and Aprijanto (2025) Machine learning method with random forest model for flood hydrometeorological disaster mitigation in Demak. International Journal on Advanced Science, Engineering and Information Technology (IJASEIT), 15 (5). pp. 1475-1484. ISSN 2088-5334

Text 02916240420261738253146.pdf Available under License Creative Commons Attribution Share Alike. Download (1MB)

Abstract

Floods are one of the most frequent hydrometeorological disasters in Demak Regency, Central Java, causing significant damage to infrastructure, livelihoods, and community safety. Given the region’s low elevation and strong seasonal rainfall patterns, this study aims to develop a localized and accurate flood prediction model based on hydrometeorological parameters. The objective is to support early warning systems and enhance mitigation planning. This research employs a machine learning approach using the Random Forest Model (RFM) and compares its performance with the Long Short-Term Memory (LSTM) model. The input data includes monthly rainfall and the number of rainy days recorded from 2022 to 2024. The study begins with data preprocessing, labelling flood categories (Major, Moderate, Low, No Flood), and model training. The dataset was split into 80% training and 20% testing sets. The RFM model achieved an average accuracy of 93.33%, precision of 91.5%, and F1-Score of 92.33%, outperforming the LSTM model, which achieved 91.67% accuracy, 90.67% precision, and 90.67% F1-Score. This suggests that RFM is more effective in classifying flood risk levels within Demak’s environmental context. In conclusion, the integration of hydrometeorological data with machine learning methods, particularly RFM, offers a reliable tool for flood disaster mitigation. The findings support the development of data-driven early warning systems and adaptive infrastructure planning. This model can serve as a decision-support system for local governments in managing future flood risks under climate uncertainty.

Item Type:	Article
Uncontrolled Keywords:	Demak, Disaster, Flood, Hydrometeorological, Random forest
Divisions:	Faculty of Information and Communication Technology
Depositing User:	Sabariah Ismail
Date Deposited:	13 Jul 2026 07:49
Last Modified:	13 Jul 2026 07:49
URI:	http://eprints.utem.edu.my/id/eprint/29996
Statistic Details:	View Download Statistic

Actions (login required)

View Item