An Integrated K-means based Data Mining Approach for Heart Disease Prediction

Abstract

Heart disease is one of the major cause of death throughout the world today. Doctors and experts shortage requires some advanced tools that can diagnose accurately and on time diagnosis of the heart disease and can give valuable information. So efficient process of extracting hidden approach patterns from massive data sets is called Data mining. Data mining also aims to identify knowledgeful patterns in the cardiac patient data.This study paper's main objective is to develop a model that provides a highly accurate forecast of heart disease. Many data mining approaches have been used in healthcare systems in the past, but hybridization—which uses more than one technique—shows promising results in identifying heart illness and can be helpful in future research on heart disease treatment. To get comparatively higher prediction accuracy, we use K-means clustering on preprocessed data. This study offers an improved prediction method with increased accuracy in addition to evaluating several recent methods for predicting heart disease. Because we combined the ID3 and C4.5, Recall and Prediction accuracy is increased effectively, the investigation created a hybrid model, which it named as MFI model in order to forecast results with high accuracy.