Feature Optimization Based Heart Disease Prediction using Machine Learning
Dhirendra Prasad Yadav, Prabhav Saini, Pragya Mittal
Abstract
Heart disease is a spontaneous, treacherous, and fatal disease. It is a group of several states that result in abnormal functioning of the heart. Based on the several pathology test report heart disease is identified by a doctor. The manual heart disease prediction is time consuming and error prone. Therefore, in the present study an automated system based on the performance analysis of several machine learning techniques has been developed. First, the well-known machine learning algorithm Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Naïve Bayes and Random Forest applied on the dataset for the prediction of heart disease. To avoid bias performance 3-fold cross validation is applied. The highest average accuracy of 87.78 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup> is obtained by the Naïve Bayes. The performance of the model is acceptable. Further, we have applied genetic algorithm on the dataset to optimize the features. After, optimization the highest average accuracy of 96% is achieved by the naïve Base.