English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Accurate epidemic forecasting plays a vital role for governments indeveloping effective prevention measures for suppressing epidemics. Most of thepresent spatio-temporal models cannot provide a general framework for stable,and accurate forecasting of epidemics with diverse evolution trends.Incorporating epidemiological domain knowledge ranging from single-patch tomulti-patch into neural networks is expected to improve forecasting accuracy.However, relying solely on single-patch knowledge neglects inter-patchinteractions, while constructing multi-patch knowledge is challenging withoutpopulation mobility data. To address the aforementioned problems, we propose anovel hybrid model called Metapopulation-based Spatio-Temporal AttentionNetwork (MPSTAN). This model aims to improve the accuracy of epidemicforecasting by incorporating multi-patch epidemiological knowledge into aspatio-temporal model and adaptively defining inter-patch interactions.Moreover, we incorporate inter-patch epidemiological knowledge into both themodel construction and loss function to help the model learn epidemictransmission dynamics. Extensive experiments conducted on two representativedatasets with different epidemiological evolution trends demonstrate that ourproposed model outperforms the baselines and provides more accurate and stableshort- and long-term forecasting. We confirm the effectiveness of domainknowledge in the learning model and investigate the impact of different ways ofintegrating domain knowledge on forecasting. We observe that using domainknowledge in both model construction and loss functions leads to more efficientforecasting, and selecting appropriate domain knowledge can improve accuracyfurther.

MPSTAN: Metapopulation-based Spatio-Temporal Attention Network for Epidemic Forecasting