A computational framework to solve the nonlinear dengue fever SIR system.

Area of Science:

• Mathematical modeling
• Epidemiology
• Computational intelligence

Background:

• Dengue fever poses a significant global health challenge, necessitating accurate predictive models.
• The Susceptible-Infected-Recovered (SIR) model is a fundamental epidemiological tool.
• Nonlinear dynamics in SIR models require advanced numerical techniques for accurate simulation.

Purpose of the Study:

• To present a numerical solution for the nonlinear dengue fever SIR system.
• To apply artificial neural networks with Levenberg-Marquardt backpropagation (ANNs-LMB) for this purpose.
• To evaluate the accuracy and reliability of the proposed ANNs-LMB method.

Main Methods:

• The ANNs-LMB technique was employed for numerical simulation.
• Data was partitioned into training (75%), validation (15%), and testing (10%) sets.
• A reference dataset was generated using the Adams scheme for comparison.

Main Results:

• The ANNs-LMB method successfully reduced the mean squared error in predicting dengue fever dynamics.
• The model demonstrated high accuracy, reliability, and effectiveness.
• Performance was validated through fitness landscapes, correlation, error histograms, and regression analysis.

Conclusions:

• ANNs-LMB provides an effective and precise numerical approach for the nonlinear dengue fever SIR model.
• The method shows strong potential for epidemiological forecasting and public health interventions.
• Validation metrics confirm the robustness and competence of the ANNs-LMB technique.