Faculty of Public Health - Andalas University - OCS, 13th IEA SEA Meeting and ICPH - SDev

Font Size: 
Artificial Intelligence based ensemble model for prediction of heart disease
nilima nilima

Last modified: 2018-09-16

Abstract


Background: heart disease is the leading cause of mortality among men and women. Accurate and timely diagnosis and detection of heart disease will assist in saving many lives.

Objective: To develop a heart disease prediction system using a novel ensemble framework based on heterogeneous classifiers namely Support Vector Machine, Naïve Bayes, and Artificial Neural Networks. The present study also verifies the most accurate algorithm among all the four.

Methodology: The data is collected from the UCI machine learning repository. After pre-processing, the data were divided into training and test data in a ratio of 80:20. We train the algorithms by providing the heart disease status. Data on disease status was available as diseased and non-diseased. Majority voting method was used to obtain the prediction results from the ensemble model.

Result: The ensemble model was observed to predict the heart disease with an accuracy of 87.05% followed by ANN (84.74%), NB (81.35%) and SVM (79.66%).

Conclusion: The ensemble model performs better than ANN, NB, and SVM. Among all individual classifiers, ANN was observed to have the least miss-classification rate. Use of the proposed ensemble classifier is recommended to predict the heart condition with better accuracy and least misclassification.