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

Table 5

Comparison of existing systems with proposed model.

Source	Technique used	Matrices used for evaluation
		Accuracy (%)	Precision (%)	Recall (%)	F1-measure (%)	ROC curve (AUC) (%)
Shafenoor Amin et al. [6] (with features engineering)	Vote	87.41	79.41		78.10
C Beulah et al. [29] (with features engineering)	Ensemble learning	85.48
Raihan et al. [30] (with features engineering)	Artificial neural network	84.47	79.97	82	85.17
Paul, Shill et al. [31] (with features engineering)	Neural network with fuzzy logic	80
El-bialey et al. [32] (with features engineering)	Decision tree	78.54
Subanya et al. [33] (with features engineering)	SVM	86.76
Djerioui et al. [34] (with features engineering)	Neighborhood component analysis and support vector machines	85.43
Proposed work (without features engineering)	SMOTE based artificial neural network	96.1	95.7	95.7	96	100