Table 2.

The results of performance of machine learning classifiers using a single questionnaire

	SVM-RBF	SVM-Lin	SVM-Poly	LDA	RF	kNN-2	kNN-5	kNN-10	LR
BDI
Accuracy	0.693	0.682	0.663	0.680	0.662	0.629	0.665	0.673	0.713
AUC	0.764	0.751	0.709	0.741	0.701	0.660	0.713	0.728	0.785
Sensitivity	0.688	0.672	0.645	0.670	0.659	0.605	0.635	0.641	0.735
Specificity	0.707	0.704	0.703	0.701	0.678	0.703	0.750	0.761	0.695
HADS
Accuracy	0.697	0.695	0.669	0.703	0.673	0.615	0.628	0.653	0.705
AUC	0.754	0.774	0.722	0.784	0.746	0.637	0.695	0.736	0.771
Sensitivity	0.698	0.714	0.661	0.710	0.675	0.589	0.601	0.620	0.719
Specificity	0.701	0.683	0.687	0.700	0.694	0.722	0.729	0.767	0.694
PHQ-9
Accuracy	0.680	0.680	0.649	0.675	0.666	0.610	0.640	0.663	0.680
AUC	0.734	0.752	0.679	0.756	0.725	0.623	0.675	0.732	0.743
Sensitivity	0.679	0.677	0.641	0.664	0.659	0.587	0.611	0.629	0.688
Specificity	0.687	0.691	0.669	0.696	0.697	0.704	0.735	0.773	0.676

BDI, Beck Depression Inventory; HADS, Hospital Anxiety Depression Scale; PHQ-9, The Patient Health Questionnaire-9; AUC, area under the curve; SVM-RBF, support vector machine-radial basis function; SVM-Lin, support vector machine-linear kernel; SVM-Poly, support vector machine-polynomial kernel; LDA, Linear Discriminant Analysis; RF, Random Forest; kNN-n, k-Nearest Neighborhood with k value of n; LR, Logistic Regression.