EFFECTIVE PREDICTION OF CARDIOVASCULAR DISEASE USING CLUSTER OF MACHINE LEARNING ALGORITHMS

Cardiovascular diseases are one of the diseases that account for the loss of millions of lives each year. Lack of early prediction is the primary reason for the loss of lives, and this encourages researchers to de-velop intelligent systems for better prediction. In this paper, a novel ensemble methodology is introduced which uses the voting of Logistic Regression(LR), Random Forest(RF), Artificial Neural Network activated with ReLU function(NNR), K-Nearest Neighbors (KNN) and Gaussian Naive Bayes(GNB) to predict the possibility of heart disease. The model is developed using Python-based Jupyter Notebook and Flask and is trained using the standard dataset from Kaggle. The model is tested and evaluated based on accuracy, precision, specificity, sensitivity, error. Testing witnessed an accuracy of 89% and a precision of 91.6%, along with a sensitivity of 86% and specificity of 91%. The results upon comparison with the individual models witness the better accura-cy of using ensemble modeling and hence a better prediction leading to life-saving.