Classification of stress of automobile drivers using Radial Basis Function Kernel Support Vector Machine

Classification of stress is imperative especially with regard to automobile drivers since stress level of the driver forms a major factor for accidents. This paper deciphered the classification of stress of automobile drivers using Radial Basis Function Kernel Support Vector Machine (SVM) classifier. The nonlinear separation of features in feature space was deciphered by this kernel trick. Pertinent feature extraction was done from ECG and EMG signals of the driver. Features extracted intuitively showed correlation with stress. This was made solid after getting a high classification accuracy of 100% using SVM using 10 fold cross validation. SVM performance was compared with that of kNN classifier and cross validation showed that kNN had only 81.26, 62.13 and 88.93% of classification rate, sensitivity and specificity where for SVM these parameters were 100%.