A Hybrid Machine Learning Model for AccurateCardiovascular Disease Prediction

Abstract

Cardiovascular diseases cause death in the world. High blood pressure, family history, stress, age, sex, BMI, cholesterol, and unhealthy lifestyle are the risk factors. In this case we introduce MaLCaDD that is a cardiovascular disease detection system, which is implemented on the machine learning base, applied feature selection, Lasso based, involves Logistic Regression, SVM, Random Forest, Decision Tree and KNN classifiers and solves the missing data and the imbalance amongst the classes. The relative analysis indicates that MaLCaDD is more accurate compared to other technologies of detecting cardiovascular disease at its earliest stage