đ
21 November 2025
DOI: 10.51903/6nagaj85
Development of An Adaptive Hybrid Weather Prediction Model Based On Pattern Classification and Deep Learning For Disaster Mitigation in Indonesia
JUISI : Jurnal Ilmiah Sistem Informasi
Universitas Sains dan Teknologi Komputer
đ Abstract
This study aims to develop and evaluate an adaptive hybrid weather prediction model that combines pattern classification techniques with a deep learning approach to improve forecasting accuracy, especially for extreme weather events. Using a quantitative-based Research and Development (R&D) approach, this study utilizes ten years of daily rainfall time series data from the Juanda Meteorological Station. The method developed comprises three main phases: weather pattern classification using K-Means clustering to separate normal and extreme patterns; development of a specialist prediction model using SARIMA for seasonal patterns and LSTM for non-linear patterns; and integration of both models into a single adaptive framework. The results show that the adaptive hybrid model performs significantly better than the single model, with a Mean Absolute Percentage Error (MAPE) of 8.76% and a Root Mean Square Error (RMSE) of 9.13%. The main contribution of this study is the development of an intelligent, accurate prediction framework with strong potential for integration into the national early warning system, thereby supporting more effective disaster mitigation efforts in Indonesia. Further research is recommended to validate the model in various regions and add additional climate variables to improve prediction accuracy.
đ Keywords
#Hybrid Model; Weather Prediction; Pattern Classification; Deep Learning; Disaster Mitigation; Model Hibrida; Prediksi Cuaca; Klasifikasi Pola; Pembelajaran Mendalam; Mitigasi Bencana
âšī¸ Informasi Publikasi
Tanggal Publikasi
21 November 2025
Volume / Nomor / Tahun
Volume 4, Nomor 3, Tahun 2025
đ HOW TO CITE
Drilanang, Mhd Ilyasyah; Indra, Zulfahmi; Walidin, Adamsyach Prana; Zai, Tri Sapta Warman ; Drilanang, Mhd Ilyasyah; Indra, Zulfahmi; Walidin, Adamsyach Prana; Zai, Tri Sapta Warman , "Development of An Adaptive Hybrid Weather Prediction Model Based On Pattern Classification and Deep Learning For Disaster Mitigation in Indonesia," JUISI : Jurnal Ilmiah Sistem Informasi, vol. 4, no. 3, Nov. 2025.
ACM
ACS
APA
ABNT
Chicago
Harvard
IEEE
MLA
Turabian
Vancouver
Download Citation
0
Sitasi
Cek Google Scholar