Heart Disease Prediction Model

Therefore, an elevated level of heart disease is a component in this examination, with 17.9million being the number of yearly fatalities that it brings about. With the number of people in the world swelling, early diagnosis and treatment are missions impossible. Notwithstanding, the recent advancements in machine learning (ML), have revolutionised the quest for health research. This study mainly concentrates on constructing a Machine learning model that can predict heart disease using some crucial parameters. Random Forest, Support Vector Machine (SVM), Naive Bayes and Decision Tree were the ML algorithms used, The Kaggle dataset which has 14 heart disease-related features was ran through these different algorithms to know which prediction technique is best for this scenario. The study also analyzes the relationship among different dataset features along with more precise predictions. From the reliable results, one of tested models was Random Forest that fitted accurately and required shorter time to process. The model was evaluated on a Kaggle dataset with 70,000 instances to strengthen the encouraging results of the study. We divided the dataset into 80:20 for training and testing, and we evaluated Decision Tree (DT), XGBoost, Random Forest (RF), Multilayer Perceptron(MP) models. Results obtained with the above set of hyperparameters: Decision Tree 85.1%, XGBoost 85.9%, Random Forest 99.7% with. This ML model can serve as useful decision support tool for physicians to predicate heart disease with more accuracy.