EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Learning from pandemics: Feature selection and clustering for robust digital disease nowcasting

Sara Mesquita, João Loureiro, Cláudio Haupt Vieira, Lília Perfeito iD, Joana Gonçalves-Sá iD

PLOS One · 2026

DOI: 10.1371/journal.pone.0358063 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2026. Verfasst von Sara Mesquita und 4 weiteren Autor:innen. Beteiligte Einrichtung: LIP - Laboratory for Instrumentation and Experimental Particle Physics, Avenida Prof. Gama Pinto 2, Lisboa, Portugal. Hauptthemen laut MEDLINE: Pandemics, COVID-19, Influenza, Human. Die Arbeit nennt 75 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação para a Ciência e a Tecnologia, European Commission.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Disease now-casting increasingly relies on non-traditional data, such as online activity, which offer fast, inexpensive insights but are often noisy, unstructured, and influenced by behavioral confounders. Models using search engine data have frequently suffered from overfitting and unstable performance. However, pandemics may help distinguish true disease signals from behavioral noise, as both tend to surge during outbreaks. We propose a hybrid approach to improve prediction that combines model evaluation with behavioral insight. We apply simple supervised and unsupervised methods to different flu and COVID-19 datasets, using the 2009 H1N1 and 2020 COVID-19 pandemics as training periods. In both cases, selecting a curated subset of search terms improved model performance, reduced overfitting, and increased stability, consistently outperforming models that used search terms indiscriminately. These results suggest that feature curation informed by behavioral insights can enhance now-casting beyond purely statistical approaches. Our method offers a scalable and adaptable framework for integrating digital epidemiology into public health surveillance.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PandemicsCOVID-19Influenza, HumanHumansClustering AlgorithmsCluster AnalysisSARS-CoV-2Influenza A Virus, H1N1 SubtypeDigital Health

Autor:innen und Institutionen

  • LIP - Laboratory for Instrumentation and Experimental Particle Physics, Avenida Prof. Gama Pinto 2, Lisboa, Portugal.
  • NOVA Medical School, Universidade NOVA de Lisboa, Campo dos Mártires da Pátria 130, Lisboa, Portugal.
  • NOVA School of Business and Economics, Universidade NOVA de Lisboa, Rua da Holanda, Carcavelos, Portugal.
  • NOVA LINCS, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa, Caparica, Portugal.

ORCID-Profile: Lília Perfeito, Joana Gonçalves-Sá

Förderung

  • Fundação para a Ciência e a Tecnologia
  • European Commission

Zitieren

APA

Mesquita, S., Loureiro, J., Haupt Vieira, C., Perfeito, L., & Gonçalves-Sá, J. (2026). Learning from pandemics: Feature selection and clustering for robust digital disease nowcasting. PLOS One. https://doi.org/10.1371/journal.pone.0358063

Vancouver

Mesquita S, Loureiro J, Haupt Vieira C, Perfeito L, Gonçalves-Sá J. Learning from pandemics: Feature selection and clustering for robust digital disease nowcasting. PLOS One. 2026. doi:10.1371/journal.pone.0358063

BibTeX
@article{Mesquita2026,
  author = {Mesquita, Sara and Loureiro, João and Haupt Vieira, Cláudio and Perfeito, Lília and Gonçalves-Sá, Joana},
  title = {Learning from pandemics: Feature selection and clustering for robust digital disease nowcasting},
  journal = {PLOS One},
  year = {2026},
  doi = {10.1371/journal.pone.0358063}
}
Kostenlos starten →