Patient satisfaction is a key quality indicator in healthcare, influencing the adherence to treatment and the overall perception of the service. However, analyzing satisfaction is complex due to multiple factors, such as quality of care, waiting times, and the environment. This study introduces an optimized Random Forest (RF) model to classify satisfaction based on survey data from 1,000 users in a public hospital in Pakistan. The analysis focuses on registration processes, medical care, laboratory and pharmacy services, and hospital environment quality. The methodology identifies influential variables that affect satisfaction and achieves the precision of the prediction 87%. It involves transforming the target variable to classify satisfaction into three levels: low, medium and high; normalizing predictors; balancing classes using SMOTE; optimizing RF hyperparameters; and extracting decision rules. The key factors identified include the attitude of the medical personnel, waiting times, availability of medication, and hospital cleanliness. The proposed model improves understanding of key drivers of patient satisfaction and offers healthcare managers a practical tool to improve service delivery effectively. By addressing the most critical aspects, this approach supports targeted interventions to improve patient experiences and overall service quality.