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

Table 11

Results of diagnosing heart disease as the category using five machine learning algorithms.

SN	Classifiers	Division of data	Class	Precision (%)	Recall (%)	F1 score (%)	Number of patients
1	SVM	Training (80%)	0	93	90	92	109
2			1	92	95	93	133
3		Testing (20%)	0	87	93	90	29
4			1	93	88	90	32
5	KNN	Training (80%)	0	88	84	86	109
6			1	88	90	89	133
7		Testing (20%)	0	90	90	90	29
8			1	91	91	91	32
9	Decision Tree	Training (80%)	0	100	100	100	109
10			1	100	100	100	133
11		Testing (20%)	0	78	86	82	29
12			1	86	78	82	32
13	Random Forest	Training (80%)	0	100	100	100	109
14			1	100	100	100	133
15		Testing (20%)	0	83	86	85	29
16			1	87	84	86	32
17	Logistic Regression	Training (80%)	0	88	83	86	109
18			1	87	91	89	133
19		Testing (20%)	0	87	90	88	29
20			1	90	88	89	32