SVM Based Risk Estimation in Heart Disease Prediction

Effective prediction models are required to identify those at risk since heart disease tends to be one of the main causes of death globally. In this work, the use of Logistic Regression (LR), Random Forest (RF), decision Tree (DT), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Ensemble Learning models for heart disease prediction is examined. The patient characteristics included in the dataset span a wide range of demographic, clinical, and lifestyle variables. To evaluate the performance of the models, performance assessment measures including recall, accuracy, and precision are used, as well as the F1-score. We have also used cross-validation score for the KNN model. To ascertain the relative impact of several variables in predicting heart disease, feature importance analysis is done. The outcomes show that all models have good prediction ability. But we can clearly see that Random Forest, Support Vector Machine, ensemble, k-Nearest Neighbors are muchbetter ML models for heart disease prediction when compared to Logistic Regression and Decision Tree models. Furthermore, we can also say that outof these techniques, SVM is the most reliable method currently that should be used by the industry.