A variant-informed decision support system for tackling COVID-19: a transfer learning and multi-attribute decision-making approach

View article
PeerJ Computer Science

Main article text

 

Introduction

  • I)

    Insufficient attention has been given to the separate consideration of data and decision support systems specifically tailored to VOC. A thorough investigation into each variant can yield valuable insights, enabling the development of more effective policies to address future VOC.

  • II)

    There is a deficiency in the assessment of each country’s performance in managing different VOCs, and a comparative analysis with other nations is lacking. Evaluating the effectiveness of policies across countries for each VOC can offer valuable benchmarks and facilitate informed decision-making.

  • III)

    There is a notable gap in leveraging prior results related to VOCs for future forecasting. By analyzing similarities between different VOCs, the utilization of past VOCs’ outcomes can be instrumental in enhancing the accuracy of forecasting the spread of future VOCs. This approach can contribute to a more comprehensive and proactive strategy in managing the ongoing challenges posed by the evolving landscape of COVID-19.

  • A Variant-Informed Decision Support System (VIDSS) is proposed and designed to adapt dynamically to each VOC. By tailoring the decision support system to the unique characteristics of each VOC, VIDSS aims to establish more effective policies, addressing the critical need for variant-specific strategies.

  • The article employs MADM techniques to introduce a novel criterion. This criterion evaluates a country’s performance by considering its improvement relative to its past state and concurrently compares it with the improvements of other countries. This approach provides a comprehensive assessment, allowing for nuanced policy comparisons among nations facing distinct VOC challenges.

  • Leveraging transfer learning, the study utilizes knowledge from forecast models of past VOCs to enhance predictions for future VOCs. By integrating insights gained from prior VOCs, this method contributes to more accurate and informed forecasting, filling the gap in utilizing historical data for proactive decision-making in the face of evolving challenges posed by COVID-19.

Framework and preliminaries

Preliminaries

Dataset description

  • 1)

    In the COVID-19 variant dataset, all subvariants are relabeled with their corresponding major VOCs. For instance, subvariants like BA.1 are renamed to their major VOCs, such as Omicron.

  • 2)

    The COVID-19 variant dataset is streamlined to focus solely on the investigation of the five major VOC: Alpha, Beta, Gamma, Delta, and Omicron.

  • 3)

    Features in the COVID-19 world dataset are aggregated based on the year-week timeframe, aligning with the temporal resolution in the COVID-19 variant dataset.

  • 4)

    The COVID-19 world dataset is filtered to exclusively include countries present in the COVID-19 variants dataset.

  • 5)

    Both datasets are merged based on country and year-week features, fostering a comprehensive alignment of information.

  • 6)

    The merged dataset is further aggregated based on country and VOCs.

  • 7)

    Null columns are eliminated, and any missing values are imputed with the median, chosen for its robustness in handling potential outliers.

MADM techniques

Transfer learning

Proposed VIDSS

Results

Managerial insights

Conclusion and future research directions

Additional Information and Declarations

Competing Interests

Erfan Babaee Tirkolaee is an Academic Editor for PeerJ.

Author Contributions

Amirreza Salehi Amiri conceived and designed the experiments, performed the experiments, analyzed the data, performed the computation work, prepared figures and/or tables, authored or reviewed drafts of the article, and approved the final draft.

Ardavan Babaei conceived and designed the experiments, analyzed the data, prepared figures and/or tables, and approved the final draft.

Vladimir Simic conceived and designed the experiments, analyzed the data, prepared figures and/or tables, and approved the final draft.

Erfan Babaee Tirkolaee conceived and designed the experiments, analyzed the data, prepared figures and/or tables, and approved the final draft.

Data Availability

The following information was supplied regarding data availability:

The data and code are available at GitHub and Zenodo:

- https://github.com/erfanmtl/variant-informed-decision-support-system

- Salehi, A., Babaei, A., Simic, V., & Babaee Tirkolaee, E. (2024). Amir27Salehi/variant-informed-decision-support-system: Initial Release (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.12804681.

Funding

The authors received no funding for this work.

1 Citation 414 Views 30 Downloads

Your institution may have Open Access funds available for qualifying authors. See if you qualify

Publish for free

Comment on Articles or Preprints and we'll waive your author fee
Learn more

Five new journals in Chemistry

Free to publish • Peer-reviewed • From PeerJ
Find out more