Score and Correlation Coefficient-Based Feature Selection for Predicting Heart Failure Diagnosis by Using Machine Learning Algorithms

Table 15

Comparison of the performance between the proposed system and previous studies.

Previous studies	Accuracy (%)	Precision (%)	Recall (%)	F1 score (%)
Arabasadi et al. [14]	93.85	—	97	—
Maji and Arora [15]	77.4	—	77.4	—
Reddy et al. [39]	90	—	91	—
Amin et al. [40]	78.15	78.15	—	80.25
Feshki and Shijani [19]	91.94	91.9	93	—
Pouriyeh et al. [41]	77.55	77.4	83	80.1
Chicco and Jurman [42]	83.8	—	72	71.9
Proposed model first dataset for training	97.68	100	100	100
Proposed model second dataset for training	100	100	100	100
Proposed model first dataset for testing	90.16	90.26	90.38	90.32
Proposed model second dataset for testing	95	97.62	95.35	96.47