Road accidents are a pervasive global issue with profound consequences for individuals, communities, and economies. This research investigates the diverse impacts of traffic accidents on human lives, healthcare systems, and economic development. Accurate accident severity analysis is crucial for effective management and prevention. To enhance prediction accuracy, this study explores the integration of machine learning methods, including Random Forest, Support Vector Machine, K-Nearest Neighbours, and Decision Tree [1]. Utilizing the Road Traffic Accident dataset, the research focuses on feature extraction and selection, aiming to classify accident severity into three levels: minor, severe, and fatal. Despite the dataset's real-world basis and inherent imbalance, this study contributes valuable insights to the discourse on road safety and accident severity prediction.