<![CDATA[This study analyzes the influence of policy (X1), governance (X2), management (X3), and public services (X4) in the Electronic-Based Government System (EBGS) on the Government Digitalization Index. A quantitative approach was employed using multiple linear regression with Stata 17, while modern analysis was conducted using deep learning with the split-sample technique. Model validation was performed using the RMSE and MAE metrics to compare the accuracies of the traditional and modern methods. The results indicate that all variables have positive and significant effects. Public services (X4) is the most dominant variable, followed by governance (X2), management (X3) and policy (X1). The consistency of the coefficient values in the training and testing data confirmed the high accuracy of the model. Deep learning analysis strengthens the regression findings by showing that policy and governance serve as catalysts that reinforce public services, whereas management functions as a support system that ensures implementation consistency. Theoretically, the findings support the grand theory of governance, resource-based view (RBV), and public service theory, emphasizing governance, internal capacity, and quality of service as the main pillars of government digitalization. This study provides an empirical contribution to the literature on the key determinants of EBGS success and demonstrates the added value of using deep learning in public policy analysis.]]>
Copyrights © 2026
