An Efficient SMOTE-Based Deep Learning Model for Heart Attack Prediction

Table 3

Average accuracy, precision, and F1-measure obtained after experimentation.

Algorithm	Computational time in seconds	Average precision	Average recall	Average F1-measure	Average accuracy	ROC value
SMOTE-based AdaBoost	61	81.3	80.4	80.2	81	83
SMOTE-based bagging	2	81.7	81	80.6	80	75
SMOTE-based random forest	2	85.1	84.3	84.4	85	82
SMOTE-based KNN	15	82.5	81.8	81.7	82	80
SMOTE-based logistic regression	15	81.8	80.7	80.6	81	70
SMOTE based naïve Bayes	22	81.9	81.5	81.4	82	70
SMOTE-based support vector machine	14	84.5	83	82.9	82	70
SMOTE-based vote	22	82.3	81.5	81	81	69
SMOTE-based artificial neural network	69	96.1	95.7	95.7	96	100