Robust Heart Disease Prediction: A Novel Approach based on Significant Feature and Ensemble learning Model

Abstract

Among the different causes of human death, heart disease is one of the most common causes of non-communicable and silent death in the world. It is a challenge to early predict heart disease by using clinical data for better treatment. After evolving machine learning, its importance is incessantly being increased in every field of life. From the last couple of years, Machine learning is also the center of attention of researchers in field medical sciences. Researchers use different tools and techniques of machine learning for the early prediction of diseases. Essentially, heart disease prediction with available clinical data is one of the big challenges for researchers. State-of-theart results have been reported using different clinical data using different machine learning algorithms, nevertheless, there is some opportunity for improvement. In this paper, we propose to use a novel method that comprises machine learning algorithms for the early prediction of heart disease. Essentially, the aims of the paper are to find those features by correlation which can help robust prediction results. For this purpose UCI vascular heart disease dataset is used and compares our result with recently published article. Our proposed model achieved accuracy of 86 <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">.</sub> 94% which outperforms compare with Hoeffding tree method reported accuracy of 85 <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">.</sub> 43%

Ähnliche Arbeiten