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

Table 1

Diagnosing heart disease features with metrics from the Cleveland dataset.

Features	Description	Explanation	Type
Age	Patient age	Age of patient in year	Numeric
Sex		1 = male	Nominal
	Patient gender	0 = female
cp		1 = typical angina	Nominal
	Chest pain	2 = atypical angina
		3 = nonanginal pain
		4 = asymptomatic
trestbps	Patient’s blood pressure at rest (mm/Hg)	Resting blood pressure (mm/Hg)	Numeric
chol	Patient’s cholesterol (mg/dL)	Serum cholesterol (mg/dL)	Numeric
fbs		1 = Fasting blood  mg/dL	Nominal
	Patient’s blood sugar during fasting	0 = Fasting blood  mg/dL
restecg		0 = normal	Nominal
	Electrocardiographic measurement at rest	1 = ST-T wave abnormality
		2 = probable left ventricular hypertrophy
thalach	Maximum heart rates	Maximum heart rate achieved	Numeric
exang	Angina due to exercise	1 = exercise induced angina	Nominal
		0 = exercise induced no angina
Oldpeak	ST depression	ST depression induced by exercise relative to rest	Numeric
Slope		1 = upsloping	Nominal
	Slope of ST	2 = flat
		3 = downsloping
ca	Number of major vessels	Number of major vessels (0-3) colored by fluoroscopy	Numeric
thal		3 = normal	Nominal
	Blood disorder	6 = fixed defect
		7 = reversible defect
Target		0 = normal	Nominal
		1 = heart disease