Informasi Lainnya
Abstraksi
This paper proposes a dynamic systems model based on Ordinary Differential Equations (ODEs) to analyze the impact of lifestyle and sociodemographic factors on diabetes incidence. The study addresses the need for interpretable and predictive models that capture the complex interactions among factors such as physical activity, diet, age, and gender. Publicly available datasets are used in parameter estimation using Random Forest Regression (RFR), identifying key transition rates among three population compartments: Susceptible, Prediabetic, and Diabetic. The ODE simulation, solved using the fourth-order Runge–Kutta (RK4) method, is evaluated with three simulation scenarios. The baseline model achieved (R2 = 0.80), the fixed 20% disturbance achieved the best alignment (R2 = 0.93), and the random 0–20% disturbance showed slightly lower agreement (R2 = 0.91) due to fluctuating disturbance intensity. These results demonstrate that consistent behavioral improvement produces more stable and sustainable outcomes than irregular disturbance, offering interpretable and data-driven insights for preventive health strategies.<br />
<br />
Index Terms—diabetes mellitus, lifestyle and sociodemo graphic, machine learning, dynamic diabetic modeling, disturbance.
Koleksi & Sirkulasi
Tersedia 1 dari total 1 Koleksi
