Identification of significant features and data mining techniques in predicting heart disease

Abstract

Cardiovascular disease is one of the biggest cause for morbidity and mortality among the population of the world. Prediction of cardiovascular disease is regarded as one of the most important subject in the section of clinical data analysis. The amount of data in the healthcare industry is huge. Data mining turns the large collection of raw healthcare data into information that can help to make informed decision and prediction. There are some existing studies that applied data mining techniques in heart disease prediction. Nonetheless, studies that have given attention towards the significant features that play a vital role in predicting cardiovascular disease are limited. It is crucial to select the correct combination of significant features that can improve the performance of the prediction models. This research aims to identify significant features and data mining techniques that can improve the accuracy of predicting cardiovascular disease. Prediction models were developed using different combination of features, and seven classification techniques: k-NN, Decision Tree, Naive Bayes, Logistic Regression (LR), Support Vector Machine (SVM), Neural Network and Vote (a hybrid technique with Naïve Bayes and Logistic Regression). Experiment results show that the heart disease prediction model developed using the identified significant features and the best-performing data mining technique (i.e. Vote) achieves an accuracy of 87.4% in heart disease prediction.