Classifying Driver Behavior Using Machine Learning: A Simple Approach to Detect Distracted and Aggressive Driving

Abstract

This study explores the use of ML models to classify driver behavior as either Distracted or Aggressive, using data derived from real-world driving scenarios. Two ML algorithms, Random Forest (RF) and Support Vector Machine (SVM), were applied to classify driver behavior based on key features such as brake_pressure, lane_deviation, and headway_distance. The RF model outperformed the SVM model, achieving an accuracy of 95% compared to 94% for SVM. The study demonstrates that brake_pressure and headway_distance are the most important features for detecting Aggressive driving, while lane_deviation is crucial for identifying Distracted driving. The findings suggest that RF is particularly effective in handling complex, high-dimensional data, providing accurate and reliable predictions. The results contribute to the advancement of road safety technologies by enhancing the detection of unsafe driving behaviors, which can be integrated into Advanced Driver Assistance Systems (ADAS) and autonomous vehicles. Future work should focus on expanding the dataset, integrating more diverse sensor data, and exploring more complex ML models, such as deep learning, to further improve classification accuracy and real-time performance in real-world applications.