3.1 Description of the Dataset and Attributes
The dataset used in this article is taken from the UCI Repository Of Machine Learning
Databases
1 . Formally, it is named Heart Disease Dataset. The Cleveland (Cleveland
Clinic Foundation) database was selected for this research because it is a commonly
used database for machine learning researchers with comprehensive and complete
records. In this field, the dataset is a collection of medical analytical reports with a total
of 303 records with 14 medical features. The various features and their description are
shown in Table 1. Besides, the categorical feature “Class” contains whether a patient
has a presence or absence of heart disease. Its original values 1, 2, 3 and 4 were
transformed in one that is the presence (1) of heart disease.
Table 1. UCI dataset attributes detailed information
Num. Code
Feature
Type
Description
1
Age
Age
Continuous Age in years
2
Sex
Sex
Discrete
sex (1 = male; 0 = female)
3
Cp
Chest pain type
Discrete
1 = typical angina;
2 = atypical angina; 3 = non-angina
pain; 4 = asymptomatic
4
Trestbps Resting boold pressure (mg) Continuous At the time of admission in hospital
[94, 200]
5
Chol
Serum cholesterol (mg/dl)
Continuous Multiple values between [Minimum
Chol: 126, Maximum Chol: 564]
6
Fbs
Fasting bood
sugar > 120 mg/dl
Discrete
1 = yes; 0 = no
7
Restecg Resting electrocardiographic
results
Discrete
0 = normal; 1 = ST-T wave
abnormal; 2 = left ventricular
hypertrophy
8
Thalach Maximum heart rate
achieved
Continuous Maximum heart rate achieved [71,
202]
9
Exang
Exercise induced angina
Discrete
1 = yes; 0 = no
10
Oldpeak ST depression induced by
exercise relative to rest
Continuous Multiple real number values between
0 and 6.2.
11
Slope
The slope of the peak
exercise ST segment
Discrete
1 = upsloping; 2 = flat;
3 = downsloping
12
Ca
Number of major vessels (0–
3) colored by fluoroscopy
Discrete
Number of major vessels coloured by
fluoroscopy (values 0–3)
13
Thal
Exercise thallium
scintigraphy
Discrete
3 = normal; 6 = fixed defect;
7 = reversible defect
14
Class
(Target)
The predicted attribute
Discrete
0 = no presence; 1 = presence
1 Repository Of Machine Learning (UCI Databases). Heart Disease Data Set. [Online]. Available:
https://archive.ics.uci.edu/ml/machine-learning-databases/heart-disease/heart-disease.names [Accessed: June 20, 2019].
302
F. Z. Abdeldjouad et al.
The dataset used in this article is taken from the UCI Repository Of Machine Learning
Databases
1 . Formally, it is named Heart Disease Dataset. The Cleveland (Cleveland
Clinic Foundation) database was selected for this research because it is a commonly
used database for machine learning researchers with comprehensive and complete
records. In this field, the dataset is a collection of medical analytical reports with a total
of 303 records with 14 medical features. The various features and their description are
shown in Table 1. Besides, the categorical feature “Class” contains whether a patient
has a presence or absence of heart disease. Its original values 1, 2, 3 and 4 were
transformed in one that is the presence (1) of heart disease.
Table 1. UCI dataset attributes detailed information
Num. Code
Feature
Type
Description
1
Age
Age
Continuous Age in years
2
Sex
Sex
Discrete
sex (1 = male; 0 = female)
3
Cp
Chest pain type
Discrete
1 = typical angina;
2 = atypical angina; 3 = non-angina
pain; 4 = asymptomatic
4
Trestbps Resting boold pressure (mg) Continuous At the time of admission in hospital
[94, 200]
5
Chol
Serum cholesterol (mg/dl)
Continuous Multiple values between [Minimum
Chol: 126, Maximum Chol: 564]
6
Fbs
Fasting bood
sugar > 120 mg/dl
Discrete
1 = yes; 0 = no
7
Restecg Resting electrocardiographic
results
Discrete
0 = normal; 1 = ST-T wave
abnormal; 2 = left ventricular
hypertrophy
8
Thalach Maximum heart rate
achieved
Continuous Maximum heart rate achieved [71,
202]
9
Exang
Exercise induced angina
Discrete
1 = yes; 0 = no
10
Oldpeak ST depression induced by
exercise relative to rest
Continuous Multiple real number values between
0 and 6.2.
11
Slope
The slope of the peak
exercise ST segment
Discrete
1 = upsloping; 2 = flat;
3 = downsloping
12
Ca
Number of major vessels (0–
3) colored by fluoroscopy
Discrete
Number of major vessels coloured by
fluoroscopy (values 0–3)
13
Thal
Exercise thallium
scintigraphy
Discrete
3 = normal; 6 = fixed defect;
7 = reversible defect
14
Class
(Target)
The predicted attribute
Discrete
0 = no presence; 1 = presence
1 Repository Of Machine Learning (UCI Databases). Heart Disease Data Set. [Online]. Available:
https://archive.ics.uci.edu/ml/machine-learning-databases/heart-disease/heart-disease.names [Accessed: June 20, 2019].
302
F. Z. Abdeldjouad et al.