This study presents a forecasting and comparative analysis of moderate geomagnetic storms using artificial neural networks (ANNs). Four moderate geomagnetic storm events that occurred during 2022, in the ascending phase of Solar Cycle 25, are investigated. The primary objective is to identify and model precursors associated with coronal mass ejections (CMEs) to forecast the SYM-H index. A supervised machine learning approach is employed, with ANNs serving as the core predictive model. The model’s performance is evaluated using standard metrics, including mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), and the coefficient of determination (R2). To benchmark the efficacy of the proposed method, the ANN-derived forecasts are compared against observed data and the outputs of several alternative algorithms: Decision Tree Regressor, Gradient Boosting Regressor, AdaBoost Regressor, and Linear Regression. The results demonstrate that the ANN model achieves superior predictive accuracy, characterized by low error metrics and a strong correlation with observed values, outperforming the other machine learning models in forecasting the SYM-H index during moderate geomagnetic storms.