An intelligent model for early prognosis of heart illness

Cardiovascular Diseases (CVDs) had arisen as deadly disease. This was the biggest reason for the world’s enormous number of deaths over the last few eras. More accuracy, perfection and precision were needed for prediction of heart diseases at early stage. To address this issue, a predictive system was needed. In medicinal arena, machine learning could be used as a tool for analysis, discovery, and prediction of numerous syndromes. Machine learning helps to make choices and forecasts effectively from vast number of records provided by healthcare field. The intent of this paper is to pinpoint threats of heart disease and to predict heart disease using Logistic Regression algorithm. The results offer the probabilities of occurrence of heart disease based on percentage. The metrics to evaluate the model for considered dataset using contingency table are Accuracy, Precision, Recall and F1 measure. Heart illness prediction using logistic regression provided enhanced accuracy than other techniques. The datasets taken are categorized based on the constraints interrelated with heart disease. This system assesses those constraints by means of machine learning practices.