Scholarlink

<p>Artificial neural networks (ANNs) have become a powerful tool in modeling complex systems due to their ability to approximate non-linear functions with high precision. Their flexibility and adaptability make them ideal for tackling intricate problems in various fields, including epidemiology. This paper introduces a stochastic Morlet wavelet neural network (MWNN) framework, optimized via a hybrid approach combining genetic algorithm (GA) and the adaptive simulated annealing (ASA), for the simulation of the Ebola virus disease (EVD) model characterized by a non-linear incidence rate. The MWNN-GA-ASA solver is constructed to capture the intricate non-linear dynamics intrinsic to the EVD system, leveraging advanced wavelet theory and stochastic optimization. No existing studies have utilized MWNN-GA-ASA optimization techniques for solving this EVD model with Non-Linear incidence rate. The performance of designed MWNN-GA-ASA is rigorously assessed through the application of multiple error metrics, including mean absolute error (MAE), root mean square error (RMSE), and Theil’s inequality coefficient (TIC), ensuring comprehensive evaluation across various dimensions. Comparative analysis with established numerical methods is presented, with performance metrics visualized using an array of graphical tools such as box plots, histograms, and loss curves. The precision, convergence, and robustness of the MWNN-GA-ASA framework are affirmed through detailed verification and validation against reference solutions, substantiating its efficacy in resolving complex epidemiological dynamics. Statistical analysis confirms the solver’s ability to accurately model the non-linear progression of EVD using an exponential incidence function, emphasizing its value as a robust computational tool for studying infectious disease dynamics.</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

Dynamic analysis of ebola virus disease with non-linear incidence rate using morlet wavelet neural networks and hybrid optimization techniques

  • 期刊论文

  • 发表日期：

    2025年1月20日

查看全文
  • Abdul Mannan,
  • Nimra Shoket,
  • Jamshaid Ul Rahman,
  • Rongin Uwitije

摘要

Artificial neural networks (ANNs) have become a powerful tool in modeling complex systems due to their ability to approximate non-linear functions with high precision. Their flexibility and adaptability make them ideal for tackling intricate problems in various fields, including epidemiology. This paper introduces a stochastic Morlet wavelet neural network (MWNN) framework, optimized via a hybrid approach combining genetic algorithm (GA) and the adaptive simulated annealing (ASA), for the simulation of the Ebola virus disease (EVD) model characterized by a non-linear incidence rate. The MWNN-GA-ASA solver is constructed to capture the intricate non-linear dynamics intrinsic to the EVD system, leveraging advanced wavelet theory and stochastic optimization. No existing studies have utilized MWNN-GA-ASA optimization techniques for solving this EVD model with Non-Linear incidence rate. The performance of designed MWNN-GA-ASA is rigorously assessed through the application of multiple error metrics, including mean absolute error (MAE), root mean square error (RMSE), and Theil’s inequality coefficient (TIC), ensuring comprehensive evaluation across various dimensions. Comparative analysis with established numerical methods is presented, with performance metrics visualized using an array of graphical tools such as box plots, histograms, and loss curves. The precision, convergence, and robustness of the MWNN-GA-ASA framework are affirmed through detailed verification and validation against reference solutions, substantiating its efficacy in resolving complex epidemiological dynamics. Statistical analysis confirms the solver’s ability to accurately model the non-linear progression of EVD using an exponential incidence function, emphasizing its value as a robust computational tool for studying infectious disease dynamics.