Inducement of multivariate factors in cardiac disease prediction with machine learning techniques substantiated with analytics

Cardiac disorder prediction is a certain requirement for preserving the lives of millions of people suffering from cardiac problems in all ages. Machine learning is a new dimension of prediction in the field of data mining as it is incorporated with mathematical techniques and procedures to provide right insight into the accurate prediction of disease with the best outcomes. The major objective of the research is to predict the cardiac disease using multivariate factors which involve; change in heart beat during exercise, oxygen supply to heart, angina responses and heart disease history. The major features attributed to the prediction of the heart disease occurrence is identified in three levels as normal, mild and severe respectively. The indication of the heart disease levels is incorporated by the rulesets formed by the multivariate factors to form a prediction network. The prediction of the multivariate component is induced with sequential application of logistic regression and linear discriminant analysis algorithms which are based on machine learning techniques. The implementation is controlled with MATLAB design and algorithm is applied on the software to predict the levels of heart disease and report in Excel format. The Analytic is performed using sensitivity and specificity measures and the accuracy is achieved with 98.2% to achieve reliability.