A Hybrid EEMD and Deep Learning Approach for COVID-19 Active Case Forecasting in Indonesia
Social Science and Humanities Journal · 2026 · 10 (07) · 10225-10232
DOI: 10.18535/sshj.v10i07.2358 ↗Kurz gefasst
Fachartikel in der Zeitschrift Social Science and Humanities Journal, 2026. Verfasst von Irene Karijadi und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Accurate forecasting of COVID-19 active cases is essential for supporting public health planning and resource allocation during infectious disease outbreaks. Nevertheless, the highly nonlinear, non-stationary, and noisy characteristics of COVID-19 time-series data present significant challenges for conventional forecasting techniques. This study proposes a hybrid forecasting framework that integrates Ensemble Empirical Mode Decomposition (EEMD) with a Long Short-Term Memory (LSTM) neural network to improve the prediction accuracy of COVID-19 active cases across six provinces in Indonesia. In the proposed approach, EEMD is first employed as a preprocessing technique to reduce noise while preserving the underlying temporal patterns of the original data. The denoised time series are subsequently used as input to the LSTM model for forecasting future active cases. The effectiveness of the proposed framework is evaluated by comparing its predictive performance with several baseline forecasting models. Experimental results indicate that incorporating EEMD as a denoising step significantly enhances the forecasting capability of LSTM, resulting in more accurate and reliable predictions for COVID-19 active cases
Zitieren
APA
Karijadi, I., Sianto, M. E., Dewi, D. R. S., & Trihastuti, D. (2026). A Hybrid EEMD and Deep Learning Approach for COVID-19 Active Case Forecasting in Indonesia. Social Science and Humanities Journal, 10(07), 10225-10232. https://doi.org/10.18535/sshj.v10i07.2358
Vancouver
Karijadi I, Sianto ME, Dewi DRS, Trihastuti D. A Hybrid EEMD and Deep Learning Approach for COVID-19 Active Case Forecasting in Indonesia. Social Science and Humanities Journal. 2026;10(07):10225-10232. doi:10.18535/sshj.v10i07.2358
BibTeX
@article{Karijadi2026,
author = {Karijadi, Irene and Sianto, Martinus Edy and Dewi, Dian Retno Sari and Trihastuti, Dian},
title = {A Hybrid EEMD and Deep Learning Approach for COVID-19 Active Case Forecasting in Indonesia},
journal = {Social Science and Humanities Journal},
year = {2026},
volume = {10},
number = {07},
pages = {10225--10232},
doi = {10.18535/sshj.v10i07.2358}
}
EUVIMED