Anticipating Norovirus Surges in Pediatric Daycare Centers: A Hybrid Prophet-LSTM-GRU Forecasting Approach
Procedia Computer Science · 2026 · 283 · 2834-2843
DOI: 10.1016/j.procs.2026.06.352 ↗Kurz gefasst
Fachartikel in der Zeitschrift Procedia Computer Science, 2026. Verfasst von Cherukupalli Yashwitha Reddy und 3 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Norovirus outbreaks pose significant public health concerns, particularly in children and the eldery. Although earlier forecasting models have explored various statistical or neural approaches independently, very few have successfully combined seasonality modeling with deep temporal learning. Deep learning models often rely on unidirectional or shallow architectures, which may limit their ability to capture bidirectional dependencies that rely on past and future events such as outbreak propagation. This work proposes a hybrid forecasting architecture that combines the Prophet model to capture seasonal trends with Bidirectional Long Short-Term Memory (BiLSTM) and Gated Recurrent Unit (GRU) components for residual sequence learning. This framework allows the model to concurrently capture long-term trends, short-term dependencies, and complex non-linear outbreak patterns. The hybrid model is trained using weekly norovirus incidence data that incorporate various exogenous features such as humidity, rainfall, temperature, and daycare metrics. The proposed hybrid model resulted in an accuracy of 97.08% and an F1- score of 97.24%. The framework provides a reliable and timely tool for norovirus outbreak forecasting, allowing daycare centers and public health authorities to plan early interventions, allocate resources effectively, and reduce the virus transmission risks.
Zitieren
APA
Reddy, C. Y., Patil, S., Sooda, K., & R, S. A. (2026). Anticipating Norovirus Surges in Pediatric Daycare Centers: A Hybrid Prophet-LSTM-GRU Forecasting Approach. Procedia Computer Science, 283, 2834-2843. https://doi.org/10.1016/j.procs.2026.06.352
Vancouver
Reddy CY, Patil S, Sooda K, R SA. Anticipating Norovirus Surges in Pediatric Daycare Centers: A Hybrid Prophet-LSTM-GRU Forecasting Approach. Procedia Computer Science. 2026;283:2834-2843. doi:10.1016/j.procs.2026.06.352
BibTeX
@article{Reddy2026,
author = {Reddy, Cherukupalli Yashwitha and Patil, Seema and Sooda, Kavitha and R, Shravya A},
title = {Anticipating Norovirus Surges in Pediatric Daycare Centers: A Hybrid Prophet-LSTM-GRU Forecasting Approach},
journal = {Procedia Computer Science},
year = {2026},
volume = {283},
pages = {2834--2843},
doi = {10.1016/j.procs.2026.06.352}
}
EUVIMED