ABSTRACT
Nowadays, cardiovascular ailments are one of the major causes of death globally, taking an estimated 610,000 lives each year. One of the biggest causes of heart disease is high blood pressure, fasting blood sugar, diabetes, high cholesterol, high BMI, and high heart rate. Heart disease diagnosis is more expansive nowadays; it involves a lot of accuracy and uncertainty due to the massive amount of data, and decisions made by doctors may fail in some cases. Data mining in healthcare is an intelligent diagnostic tool. Thus, it is compulsory to predict the health risks of every human being, depending on age, sex, blood pressure, diabetes based on symptoms along with what we can do for precaution which is done by diagnosing disease and providing appropriate cure on the right instant. This chapter aims to forecast a classification model and to know which kind take part in ailment forecast by using Cleveland and Statlog heart dataset. Prediction is implemented with six techniques, naïve Bayes, k-NN, random forest, logistic regression, SVM, and decision tree, on different datasets, followed by a comparative analysis of the classification model for better accuracy and results. A random forest outperformed the other classifiers with an accuracy rate of 99.80%, followed by logistic regression.
