Optimized Machine Learning Model for Driver’s Drowsiness Detection Using KNN Classifier

Optimized Machine Learning Model for Driver’s Drowsiness Detection Using KNN Classifier

Meghana, Pooja H S, Sanjana E, Vinutha Shetty E, Sheela B P

Meghana Information Science & Engineering& RYMEC Ballari

Pooja H S Information Science & Engineering& RYMEC Ballari

        Sanjana E Information Science & Engineering& RYMEC Ballari                                                                                                       Vinutha Shetty E Information Science & Engineering& RYMEC Ballari                                                                                          Sheela B P Assistant Professor Dept of ISE RYMEC Ballari

---------------------------------------------------------------------***--------------------------------------------------

Abstract - This study explores using machine learning, specifically the K-Nearest Neighbors (KNN) algorithm, to detect driver drowsiness, a major cause of traffic accidents.  The research focuses on optimizing the KNN model for accurate and efficient real-time detection.  This involves selecting the most informative features from a dataset of driver behavior and physiological signals (like eye movements and facial expressions), tuning the KNN algorithm's parameters for best performance, and possibly reducing the data's complexity.  The optimized model's performance is tested, aiming to create a reliable system that can be integrated into vehicles to prevent accidents by alerting drowsy drivers.

Key Words:  KNN, Feature Selection, Parameter Tuning, Machine Learning.