Digital Video Broadcasting (DVB) systems require robust signal quality monitoring to ensure optimal performance. This study presents a machine learning-based framework for predicting Bit Error Rate (BER) and classifying signal quality (Poor, Moderate, Good) using key transmission parameters such as Signal-to-Noise Ratio (SNR) and received power. We employ Random Forest regression for BER estimation and Random Forest classification for quality assessment, leveraging feature engineering to derive meaningful relationships between signal characteristics. Our approach includes outlier detection, cross-validation, hyperparameter tuning, and performance evaluation using metrics such as RMSE, MAE, R2 (for regression) and accuracy, precision, and confusion matrix analysis (for classification). The results demonstrate strong predictive capability, with feature importance analysis revealing SNR as the most critical factor in signal quality determination. This work provides a data-driven solution for proactive DVB network monitoring, enabling early fault detection and performance optimization.