Multi-Feature Representation Based COVID-19 Risk Stage Evaluation With Transfer Learning

Xiangjie Kong, Ning Li, Chenwei Zhang, Guojiang Shen, Zhaolong Ning, Tie QiuIEEE Trans. Network Science & Engineering 2022 · OpenReview

Abstract

Accurate real-time COVID-19 confirmed case prediction and risk stage evaluation are of great significance for government decision-making. However, the complexity of the epidemic spread and the lack of data in countries where COVID-19 has recently emerged are still an important challenge for researchers. In this paper, we propose a multi-feature representation based COVID-19 risk stage evaluation model with transfer learning (COV2RS) to solve these issues. Specifically, by analyzing COVID-19 spatio-temporal characteristics, we build a spatial epidemic spread network and obtain multi-feature representation using a hierarchical spatio-temporal framework. From this, we obtain the COVID-19 confirmed case prediction results by aggregating representations. Finally, with transfer learning, we utilize COVID-19 confirmed data from other countries to evaluate the risk stage. Experiments based on real-world COVID-19 datasets demonstrate that our prediction results outperform the state-of-the-art baselines and can evaluate the COVID-19 risk stage well.
BibTeX
@article{kong2022multi,
  author = {Xiangjie Kong and Ning Li and Chenwei Zhang and Guojiang Shen and Zhaolong Ning and Tie Qiu},
  title = {{Multi-Feature Representation Based COVID-19 Risk Stage Evaluation With Transfer Learning}},
  journal = {IEEE Trans. Network Science \& Engineering},
  year = {2022},
  url = {https://openreview.net/forum?id=qeS8yxcUZvV},
}

Loading related content...

Figure