Predicting Accurate Heart Attacks Using Logistic Regression

Abstract

A heart attack is one of the leading causes of death today. According to a large data population used as a training set for the algorithm for machine learning, classification is a technique for predicting the target class from input data. A difficulty in clinical data analytics is predicting heart attacks with greater precision. The focal point of this work is to analyze the heart attack dataset (Kaggle repository) to find a Machine learning classifier technique that predicts if a person is prone to a heart attack with maximum accuracy based on various health factors. The efficacy of the three classifiers, namely Logistic Regression, Random Forest, and Decision Tree, is demonstrated for predicting heart attack. This work compares the three classification algorithms among various factors. Logistic Regression outperforms all for predicting the values from the dataset accurately.