Enhanced Heart Disease Prediction Using Ensemble Learning Methods
Number of people losing their lives due to heart disease is growing day by day, revealing the need of a model which predicts beforehand. An initiative has to be taken to aid the people by giving them a cautionary advice about the disease at the correct time. It is not easy for everyone to afford expensive treatments and medications so there is urgency of a structure which can quickly go through the information of the patient and inform them at an earlier stage if they test positive. We need a logical process that analyzes and finds unrevealed data and figures in the medical data. Thus we propose to perform the analysis of given dataset by performing the data validation and preprocessing techniques, exploration data analysis visualization and training a model, build a classification model and then performance measurements of supervised machine learning algorithms with evaluation classification report, identify the confusion matrix and categorizing data from priority. The main objective is to make a predictive analytics model to diagnose the various stages of heart patients by ensemble learning methods like Bagging, Boosting and Voting which aims to enhance the accuracy of the deficient algorithms. Outcome of these ensemble techniques are analyzed and the one that proves to enhance the precision is considered and showcased using a GUI.