ABSTRACT
Heart disease, often recognized as a cardiovascular disease that, denotes a range of conditions that grief the heart and has become the prominent cause of death worldwide in recent decades. It combines multiple cardiovascular disease risk variables with the requirement for time to produce precise, accurate, and sensitive methods for early identification and treatment of the illness. In the realm of healthcare, data mining is a commonly utilized tool for evaluating large amounts of data. The search employs a variety of data mining and machine-learning approaches to evaluate massive amounts of complex medical data to assist health providers in predicting cardiac issues. This research comprises a model based on supervised learning algorithms such as Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Random Forest algorithm and several factors linked to heart disease. It makes use of the current Cleveland database, which is made up of persons with heart disease from the heart disease dataset available on Kaggle. The goal of this work is to figure out how likely it is that a patient would acquire heart disease. The outcome of the proposed system reveals that the greatest accuracy score is achieved with the Random Forest algorithm in a desktop application form. After implementing three approaches, the system discovered that the accuracy in the Random Forest was the highest 114(96%). The research will be enhanced with various machine learning (ML) methods, grouping, and association rules, vector machine assistance, and evolutionary algorithms.
