Utilizing Machine Learning for Heart Disease Prediction

Document Type : Original Article

Authors

1 College of Technology and Applied Sciences- Al Quds Open University Bethlehem-Palestine

2 College of Technology and Applied Sciences- Al Quds Open University Ramallah-Palestine

3 College of Technology and Applied Sciences- Al Quds Open University Jenin-Palestine

Abstract

Over the past few year, the world has witnessed a significantly increase during heart disease prevalence, which threatens the safety of people's lives, and it has become one of the most common diseases today. Consequently, finding the most effective method for early disease prediction is essential, as it can improve lives.
This study seeks to design a system for predicting heart diseases that enables them to predict the probability of a person suffering from a heart disease in order to prevent him from it, based on the medical history of the patient.
We used different ML algorithms available in WEKA version 3.8.1. Examples include Naive Bayes, logistic regression, and decision tree J48 for classification of patients with heart disease
It includes 3 steps: First select data for 18 clinical features from a kaggle-like site BMI, Smoking, AlcoholDrinking, Stroke, PhysicalHealth...etc. Second, initialize the data, Thirdly, the development of the tree algorithm, Naive bayes, and logistic regression to predict Heart disease determined by clinical characteristics.
The logistic regression model achieved an accuracy of 92.1%, which is significantly higher than other algorithms tested, as it effectively predicted the likelihood of heart disease in individuals.
The heart disease prediction system provides health care to save human life, using appropriate medications.

Keywords

Files

Share

How to cite

Statistics