International Journal of Computer Techniques Volume 12 Issue 4 | A Machine Learning Approach to Driver Behavior Classification Using Telematics Data
A Machine Learning Approach to Driver Behavior Classification Using Telematics Data
Authors:
Karinigam S A, Final Semester MCA Student, CMR University, Bengaluru, India (karinigam.sa@cmr.edu.in)
Dr. Deepa A, Associate Professor, CMR University, Bengaluru, India (deepa_a@cmr.edu.in)
Akhila S Babu, Assistant Professor, CMR University, Bengaluru, India (akhila_s@cmr.edu.in)
Journal: International Journal of Computer Techniques (IJCT)
Volume: 12 | Issue: 4 | Publication Date: July – August 2025
ISSN: 2394-2231 | Journal URL: https://ijctjournal.org/
Abstract
This paper proposes a machine learning framework for classifying driver behavior using telematics data. Four models—Random Forest, SVM, CNN, and LSTM—were evaluated on the UAH-DriveSet dataset to detect normal, aggressive, and drowsy driving. The system architecture integrates sensor data preprocessing, feature extraction, and real-time classification. Results show high accuracy and practical applicability in usage-based insurance, fleet management, and driver assistance systems.
Keywords
Driver Behavior Classification, Telematics, Machine Learning, Deep Learning, Random Forest, SVM, CNN, LSTM, Intelligent Transportation Systems
Conclusion
The study confirms the feasibility of using ML and DL models to classify driver behavior from telematics data. Random Forest provided interpretability, while LSTM captured temporal dynamics. The system supports real-time safety interventions and long-term behavioral insights. Future work should address generalization, model transparency, and privacy concerns to enable scalable deployment.
References
Includes 15 references from IEEE, Applied Sciences, and European Transport Research Review covering telematics, driving behavior analytics, and intelligent transportation systems.
