The progression of the COVID-19 pandemic has demonstrated significant oscillatory characteristics, underscoring the importance of investigating the impact of driving factors on its evolution. This study included an in-depth analysis of the influence of various driving factors on the pandemic’s fluctuations, identifying key elements, to enhance the comprehension of transmission mechanisms and improve scientific precision in formulating mitigation strategies. The experimental outcomes indicate that the Geographically and Temporally Neural Network Weighted Regression (GTNNWR) model achieved commendable accuracy with minimal error in forecasting the number of infected individuals. Leveraging the results from the GTNNWR model, the research meticulously examines the temporal and spatial correlations between the driving factors and the pandemic, delineated the spatiotemporal distribution patterns of each factor’s influence, and quantified their significance. This study reveals the substantial impact of vaccines, masks, and social distancing measures across different regions and periods, with their effects on the number of affected individuals being 2 to 10 times more pronounced than other driving factors. These findings contribute to a deeper understanding of the spatiotemporal transmission dynamics and the influence of driving factors in the COVID-19 pandemic, offering critical decision-making support for control and prevention efforts.