An Optimized Approach Using Transfer Learning to Detect Drunk Driving

Table 1

The performance of classifiers over the same dataset.

Classifier	Accuracy	Precision	Recall	F1-score	AUC	MCC
CNN	0.5441	0.5238	0.9705	0.6804	0.5441	−0.0563
CNN + KNN	0.625	0.6538	0.85	0.7391	0.55	0.1240
CNN + SVM	0.6562	0.8	0.6	0.6857	0.675	0.3395
CNN + RF	0.9375	1.0	0.9	0.9473	0.95	0.8783