Analyzing the Comparison of C4.5, CART and C5.0 Algorithms on Heart Disease Dataset using Decision Tree Method

Abstract

Data acquisition methods can be expected for patients suffering from heart disease. The resolution of this learning was to compare a similar data mining algorithm to the calculation of heart diseaseThis research paper proposed the traditional decision tree procedure and weighted decision tree procedure. Traditional decision tree process consists of C4.5, C5.0, CART processes. The weighted decision process is established suitable weights of training cases based on naïve Bayesian theorem before trying to construct a decision tree model. The main objectives of this research paper are (1) to know the operation of C4.5 process, Cart and C5.0 process, (2) to learn how to analysis the traditional decision tree and weighted decision tree algorithms are compared results from both training and testing dataset for heart disease.

A New Efficiency Improvement of Ensemble Learning for Heart Failure Classification by Least Error Boosting

“…It was found that serum creatinine and ejection fraction were the two most relevant features, and an important feature can lead to more accurate predictions than using the original dataset features in a variety of machine learning test methods. This is similar to the research by Myint & Khaung [7], which uses supervised learning in the decision tree method to classify comparisons of heart disease analysis, including C4.5, C5.0, and Cart. The results showed that the C5.0 decision tree was perfected with the greatest accuracy.…”

“…Machine learning was one of the tools that was used to analyze a large number of heart failures, using a variety of methods, but popular methods used in predicting include supervised learning, deep learning, and ensemble. In this case, supervised learning is a method for creating a model for predicting the results of the analysis or the heart failure prognosis by learning the existing patient data to make decisions on new data to analyze [4][5][6][7]. Supervised learning was used to increase the efficiency of the prediction of heart failure; for example, Chicco & Jurman [8] applied several machine learning classifiers with feature ranking tested on heart failure datasets of 299 patients.…”