Cervical cancer is a major threat to public health that continues to rank among the most

prevalent causes of cancer-related fatalities in low- and middle-income nations. This reflects

significant inequities caused by a lack of accessibility to the national HPV vaccine, cervical

screening, and medical programs, as well as economic and social factors. Therefore, this study examined the cervical cancer epidemic model with three classes that are Susceptible, HPV-infected, and HPV-induced females affected by cervical cancer, by using stochastic numerical schemes. The proposed model is numerically solved by using the Levenberg-Marquardt Backpropagated Technique (LMBT) based on neural-networks. Datasets reflecting variations in women's risk of acquiring HPV infection and the probability of cervical cancer-related deaths are generated using Adam's methods. These datasets are then utilized by neural networks to perform analyses, including histograms, regression, time series, cross-correlation, and statistical analysis, to estimate the efficiency and accuracy of LMBT. Additionally, absolute errors and mean squared errors are calculated to evaluate LMBT convergence. Following the demonstration of LMBT’s effectiveness, the study explored the dynamic behaviors of cervical cancer by increasing women's risk of acquiring HPV infection and the probability of cervical cancer-related deaths by this technique.