Exploring Machine Learning Techniques for Accurate Heart Disease Detection: A Comprehensive Study

Globally, heart disease poses a significant challenge to public health. Early identification is crucial for effectively managing and treating heart conditions. Machine learning has shown promising results in predicting medical problems for the past few years. In this study, A system was developed that is capable of detecting cardiovascular diseases using machine learning methods. The project consists of Data loading, Exploratory data analysis, preprocessing, training, evaluation and storage. A dataset was introduced that contained an array of characteristics related to the cardiovascular health status (such as age, gender, cholesterol level.). However, training measured the accuracy of 3 algorithms: K Nearest Neighbours (KNN), Decision Tree, and Random Forest in identifying heart ailment. The article also returns an ensemble graph that considers the forecasts of both these models together, saving the best GNB model for future use. Results showed that the machine learning methods are proficient in heart problem identification, as in this hybrid model, with a 97.32% accuracy rate.