Abstract
In this paper, a new Meyer neuro-evolutionary computational algorithm is introduced for mathematical modeling of the epidemiological smoking model by employing hybrid heuristics of Meyer wavelet neural network with global optimized search efficiency of genetic algorithm and sequential quadratic programming. According to the World Health Organization, tobacco consumption kills 10% of all adults worldwide. The smoking epidemic is often regarded as the greatest health threat that humanity has ever confronted. So it's an important issue to address by employing hybrid suggested techniques. The Meyer wavelet modeling approach is exploited to describe the system model epidemiological smoking in a mean squared error-based function, and the systems are optimized using the proposed approach's combined optimizing capability. Root mean square error, Theil's inequality factor, and mean absolute deviation-based measurements are used to better verify the effectiveness of the suggested methodology. The combined approach for smoking model is verified, validated, and perfected through comparison investigations of reference results on stability, precision, convergence, and reliability criteria, which shows the novelty of this study. Furthermore, the results of the quantitative study support the value of the suggested approach-based stochastic algorithm. The values of absolute error lie between 10(-2) and 10(2), 10(-6) and 10(-3), 10(-6) and 10(-4), 10(-6) and 10(-3), 10(-5) and 10(-4), and 10(-6) and 10(-5). The convergence measurement values for Theil's inequality coefficient lie between 10(-5) and 10(0), 10(-10) and 10(-5), 10(-10) and 10(-5), 10(-10) and 10(-5), 10(-10) and 10(-5), and 10(-10) and 10(-5).