A Machine Learning Approach Towards Heart Attack Prediction

Abstract

Various types of heart diseases and conditions leading to increasing chance of heart attack have been a serious concern all over the world. Several factors like blood pressure, cholesterol, diabetes, obesity can affect the heart, and thus, those should be monitored regularly to prevent the chance of heart attack in people of different age groups. This chapter at first has analyzed different existing benchmarks of heart attack analysis. Being motivated by the shortcomings of the state-of-the-art literature and to address the challenges, it has introduced support vector machine, the most popular supervised machine learning algorithm to classify the chance of heart attack using a dataset downloaded from Kaggle. The experimental result has been evaluated using different performance metrics, including accuracy, error rate, precision, recall, F1 score. Finally, the performance has been compared with the existing related works also to validate its effectiveness and efficiency in real-time heart attack prediction.