Scholarlink

<p>This research provides a computing neural network approach for solving the complex co-endemic dynamics of dengue and coronavirus, which provides a prospective operator to understand and predict the co-infection designs. A stochastic computing machine learning neural network process is designed based on a single hidden layer construction along with the activation radial basis function, twelve neurons, and training through Levenberg–Marquardt backpropagation scheme. The design of data is performed through the explicit Runge–Kutta, which decreases mean square error by separating into training as 70%, validation as 12%, and testing as 18%. The accuracy of the designed machine learning NN process is evaluated based on how well the outputs align in good order, and small absolute error found as 10<sup>–06</sup> to 10<sup>–08</sup>. The best training is measures as 10<sup>–09</sup> to 10<sup>–11</sup>, regression coefficient, error histogram also validate the preciseness of the proposed machine learning neural network.</p>

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

An optimization based neural network enhanced by radial basis for the co-endemic model of dengue and coronavirus

  • 期刊论文

  • 发表日期：

    2026年3月2日

查看全文
  • Zulqurnain Sabir,
  • Abdulrahman Al Zaatari,
  • Youssef Chehab,
  • Kamel Jaafar,
  • Mustafa Bayram,
  • M. A. Abdelkawy

摘要

This research provides a computing neural network approach for solving the complex co-endemic dynamics of dengue and coronavirus, which provides a prospective operator to understand and predict the co-infection designs. A stochastic computing machine learning neural network process is designed based on a single hidden layer construction along with the activation radial basis function, twelve neurons, and training through Levenberg–Marquardt backpropagation scheme. The design of data is performed through the explicit Runge–Kutta, which decreases mean square error by separating into training as 70%, validation as 12%, and testing as 18%. The accuracy of the designed machine learning NN process is evaluated based on how well the outputs align in good order, and small absolute error found as 10–06 to 10–08. The best training is measures as 10–09 to 10–11, regression coefficient, error histogram also validate the preciseness of the proposed machine learning neural network.