Exploration of Black Boxes of Supervised Machine Learning Models: A Demonstration on Development of Predictive Heart Risk Score

Table 3

Performance assessment of NLHRS-based risk prediction models.

Models	ANN-RS		SVM-RS
Confusion matrix	Case	Control	Case	Control
Case	190	40	190	40
Control	48	182	36	194
Accuracy	81.000		83.500
Sensitivity	0.826		0.826
Specificity	0.791		0.843
Kappa statistic	0.620		0.670
AUC	0.876		0.888
RMSE	0.378		0.362