Scholarlink

The last COVID-19 pandemic occurred in several unpredictable waves, making the classical compartmental models useless. To overcome this issue for a potential future pandemic, we have adopted a new approach based on unsupervised learning with the Gaussian Mixture Model. In this regard, we have applied our model to several datasets of infected populations from various countries around the world. The outcomes have demonstrated the flexibility and accuracy of this model in comparison to the SIR model and its derivatives. In this work, we rely on confirmed cases and death data for all countries, sourced from Our World in Data. Furthermore, our findings have led to the development of a new dashboard. Our results have been validated through statistical estimation and illustrated using the R language.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Wave-Based Pandemic Modeling Using Unsupervised Learning: Gaussian Mixture Model Approach

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Sidi Mohamed Lalaoui Ben Cherif,
  • Abdous Belkacem,
  • Abdelmajid Bessate

摘要

The last COVID-19 pandemic occurred in several unpredictable waves, making the classical compartmental models useless. To overcome this issue for a potential future pandemic, we have adopted a new approach based on unsupervised learning with the Gaussian Mixture Model. In this regard, we have applied our model to several datasets of infected populations from various countries around the world. The outcomes have demonstrated the flexibility and accuracy of this model in comparison to the SIR model and its derivatives. In this work, we rely on confirmed cases and death data for all countries, sourced from Our World in Data. Furthermore, our findings have led to the development of a new dashboard. Our results have been validated through statistical estimation and illustrated using the R language.