Detection of Arrhythmia using Ensemble classifier in Comparison with Support Vector Machine Classifier to Measure the Accuracy, Sensitivity, Specificity and Precision
Kirupa Ganapathy, P. Karthikeyan, L. Harshitha · 2022
Aim: Machine Learning is used as a technique for fully paid loan payback with an increase in accuracy of prediction utilising Support Vector Machine (SVM) in comparison to XGBoost (XGB) classifier on lending club data. The components and procedures It was decided to use a sample size of 4000 for the completely paid loan repayment system forecast, with Group 1 equaling 2000 and Group 2 equaling 2000. Support Vector Machine (SVM) and XGBoost (XGB) Classifier are two strategies that are utilised in the process of classification of enhanced fully paid loan repayment with an increase in accuracy prediction. In the end, the accuracy rate of the Support Vector Machine (SVM) was found to be 94.18%, while the accuracy rate of the XGBoost (XGB) algorithm was found to be 89.34%. When compared, the findings of XGBoost (XGB) have a precision rate of 90.21%, whereas Support Vector Machine (SVM) has a precision rate of 94.55%. When compared, the Support Vector Machine (SVM) has a recall rate of 95.62%, whereas the results of XGBoost (XGB) have a recall rate of 90.45%. The Support Vector Machine (SVM) achieves a specificity rate of 95.34%, while the XGBoost (XGB) algorithm only achieves a specificity rate of 90.97%. There is a statistically significant gap in the rate of accuracy (P = 0.062). When it comes to coming to a conclusion, the Support Vector Machine (SVM) Classifier outperforms the XGBoost (XGB) Classifier in terms of finding the better classification in terms of finding the accuracy, precision, recall, and specificity for predicting fully paid loan repayment systems.