ENHANCED WEIGHTED QUADRATIC RANDOM FOREST ALGORITHM FOR HEART DISEASE PREDICTION

Abstract

Heart disease is the leading reason for death in the U.S. Sooner or later in your life, possibly you or one of your friends and family will be forced to settle on decisions about some part of heart disease. Heart disease can strike out of nowhere and expect you to settle on decisions rapidly. In this phase, the proposed Enhanced Weighted Quadratic Random Forest Algorithm is applied to patient heart disease data with high-dimensional lopsided attributes. First, the random forest algorithm is utilized to arrange feature significance and diminish dimensions. Second, the chose features are utilized with the random forest algorithm and the F-measure esteems are determined for every decision tree as weights to construct the prediction model for patient heart disease data. Weighted F-measure into the RF algorithm, which creates a superior performance for patient heart disease prediction by assigning different weights to different decision trees with the proper version of proposed classifiers, this technique could thus build up a most ideal measure of covered up units for guaranteed.