EUVIMEDEUVIMED entdecken
Crossref

Data and model dual-driven: Precise prediction of Infectious Diseases Based on Spatio-temporal graph Neural networks

Chengyun Li

Advances in Engineering Technology Research · 10.03.2026 · 16 (1) · 317

DOI: 10.56028/aetr.16.1.317.2026 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advances in Engineering Technology Research, online erschienen am 10.03.2026. Verfasst von Chengyun Li.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurately predicting the spread of infectious diseases such as the novel coronavirus is a core challenge in public health management. Most existing models do not account for the network structure of transmission and treat different regions as isolated time series, which affects prediction performance. This paper proposes a data-driven spatio-temporal graph neural network framework, Corr-STGNN. Its core is two-stage modeling: constructing a functional association graph based on Pearson correlations of historical cases across different regions to capture potential spatial connections between non-adjacent regions, and then combining the graph convolutional network with a gated recurrent unit to learn spatial dependence and temporal dynamics jointly. Data validation across multiple states in the United States shows that this method is more accurate than the non-spatial GRU benchmark model, reducing the root mean square error by over 27%. Although Corr-STGNN overcomes the traditional reliance on static adjacency or human flow data by constructing data-driven association graphs, it still has limitations, such as being sensitive to external data and failing to capture dynamic functional connections fully.

Zitieren

APA

Li, C. (2026). Data and model dual-driven: Precise prediction of Infectious Diseases Based on Spatio-temporal graph Neural networks. Advances in Engineering Technology Research, 16(1), 317. https://doi.org/10.56028/aetr.16.1.317.2026

Vancouver

Li C. Data and model dual-driven: Precise prediction of Infectious Diseases Based on Spatio-temporal graph Neural networks. Advances in Engineering Technology Research. 2026;16(1):317. doi:10.56028/aetr.16.1.317.2026

BibTeX
@article{Li2026,
  author = {Li, Chengyun},
  title = {Data and model dual-driven: Precise prediction of Infectious Diseases Based on Spatio-temporal graph Neural networks},
  journal = {Advances in Engineering Technology Research},
  year = {2026},
  volume = {16},
  number = {1},
  pages = {317},
  doi = {10.56028/aetr.16.1.317.2026}
}
Kostenlos starten →