An Effective Heart Disease Prediction Method using Extreme Gradient Boosting Algorithm Compared with Convolutional Neural Networks

Heart disease is one of the most serious and common problems to human beings around the face;% to 90% of people are affected by heart problems in which, and Several methods exist to identify heart disease in an earlier stage, but few methods fail to manage the accuracy due to the large volume of data. In addition, the methods require more time to process those features. To overcome these research issues, effective intelligent techniques are proposed to predict heart disease in an earlier stage. Additionally, our model’s balanced accuracy (89.5%) outperforms the XGBoost classifiers' accuracy rates. Excellent recall values and area under the curve, together with high specificity and test accuracy, are obtained using the CNN classifier. Designers contrast the XGboost, KNN, SVM, and Naive Bayes Classifiers' classification precision on various healthcare datasets. Finally, this leads to a reduction in the overall error rate and also improves the recognition accuracy by more than 90%.