completed
Alcohol-Influenced Driving Detection from Smartphone Sensors
A decision-tree classifier detecting alcohol-influenced driving from smartphone sensor data collected in Nigeria.
A continuous, smartphone-sensor approach to detecting alcohol-influenced driving suited to low-resource contexts. The model was trained on a Nigeria-collected dataset and achieves recall 100%, precision 60%, F1 75%, and accuracy 90.91%.
Technology
Pythonscikit-learnDecision TreesSmartphone Sensors
Themes
Machine LearningNigeriaDriving BehaviourSmartphone Sensing