Research Article

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

Table 1

Dataset description.

Sr no.Data typeAttributesDescription

1NumericAgeIt is described in years
2BinarySexGender where 1 represents male while 0 represents female
3CategoricalChest pain (CP)It is type of chest pain having four classes with varying degree with values from (0 to 3)
4NumericResting blood pressure (testbps)Measure of blood pressure in mmHg (94 to 200)
5NumericCholestrol (chol)Measure of cholesterol in blood (126 to 554)
6BinaryFasting blood sugar (fbs)Measure of sugar in blood with fasting (0, 1)
7CategoricalResting ECG (restecg)Measure of ECG in resting phase having three classes: normal, ST defected, and ventricular hypertrophy (0, 1, and 2)
8NumericThalachMaximum heart rate in beats per minute (71 to 202)
9BinaryExangExercise-induced agina (1 and 0)
10DecimalOld peakST depression induced by exercise (0 to 4.4)
11OrdinalCaNumber of major vessels colored (0 to 3)
12CategoricalThalRepresents thallium stress test results (0 = NA, 1 = fixed defect, 2 = normal, 3 = reversible defect)
13CategoricalSlopeThe slop of ST segment having three classes, upward, flat, and downward slopes with values of 0, 1, and 2
14BinaryTargetTarget attributes having two classes where 1 indicates presence of heart attack 0 indicates absence of heart attack