Predictive Model Based on Supervised Machine Learning for Heart Disease Diagnosis

Abstract

Nowadays, heart disease has become a very common disease in human beings. Due to the rapidly increasing number of heart disease cases, it has now become inevitable to have some efficient and accurate mechanism that can detect heart disease early in patients. Although various heart disease diagnosis methods are available in modern medical science, these methods are very costly and involve a lot of analysis of the test results obtained during the patients’ health examinations. To ease the heart disease diagnosis process, there exist several supervised machine learning techniques which can predict heart disease in patients based on laboratory test data. Prediction technique that is the most efficient, accurate, and suitable over the heart disease dataset for prediction is still required to be explored. So, this research contribution focuses on identifying the potential predictive model based on supervised machine learning suitable in the prediction of heart disease over the heart disease dataset. The proposed work found Logistic Regression to be the promising predictive model with classification accuracy 85.3 %, F1- Score 80%, Precision 82.6%, and recall 85.3%.