Of the prediction model are: Accuracy = 0.87665 = 0.87665 = 0.94654 AUC = 0.The AUC is plotted
On the prediction model are: Accuracy = 0.87665 = 0.87665 = 0.94654 AUC = 0.The AUC is plotted together with the line representing the True-Positive Price of 0.five along with the AUC is plotted in addition to the line representing the True-Positive Price of 0.five and the False-Positive Price of 0.five to show the efficiency of your model, and this system on the False-Positive Price of 0.5 to show the efficiency with the model, and this approach of validation is called the ROC curve evaluation. Figure 8 shows the result on the ROC curve validation is called the ROC curve evaluation. Figure 8 shows the result of your ROC curve evaluation performed for the model educated and tested within this investigation, plus the AUC is far analysis performed for the model educated and tested in this study, and also the AUC is far away from the 0.5 line, which signifies that the model covered the dataset well and can away from the 0.five line, which implies that the model covered the dataset well and can predict the student dropout or continue for many cases within the dataset. predict the student dropout or continue for most cases inside the dataset.Figure eight. The ROC of your Model. Figure eight. The ROC of the Model.The variation of accuracy, precision, recall, and F1-score with the model for diverse The variation of accuracy, precision, recall, and F1-score with the model for various days are shown in Figure 9.9. It may be observed that the accuracythe model is regularly days are shown in Figure It might be observed that the accuracy of of your model is consistaboveabove 70 and largely above the precision of theof the model is generally above and ently 70 and mostly above 80 , 80 , the precision model is constantly above 70 70 BCECF-AM MedChemExpress consistently above 80 80 and mostly above 90 , the recall with the model is alwaysabove and regularly above and mostly above 90 , the recall with the model is normally above 80 and consistently above 90 , and also the F1-score with the model is always above 70 and regularly above 80 and mainly above 90 , respectively.TP-064 Technical Information information 2021, 12, x FOR PEER REVIEW14 ofInformation 2021, 12,80 and consistently above 90 , as well as the F1-score of your model is constantly above 70 and regularly above 80 and largely above 90 , respectively.14 ofFigure 9. (a). Accuracy with the Model on Various Days Precision of with the Model on Distinctive (c). Recall of your from the Figure 9. (a). Accuracy of the Model on Diverse Days (b).(b). Precisionthe Model on Different Days Days (c). Recall Model on Distinct Days (d). F1-score in the from the on Distinct Days. Model on Unique Days (d). F1-scoreModel Model on Different Days.These final results show that the model performs nicely for any offered set of information, because the These outcomes show that the model performs nicely for any given set of data, because the dataset has significantly less data as the number of of days increases, but this really is reflected on the perdataset has much less information because the number days increases, but this isn’t not reflected around the functionality of model, showing the robustness from the model. Nevertheless, even with these formance of thethe model, showing the robustness on the model. Having said that,even with these benefits, the model can’t be explained. Hence, this research utilizes the SHAP visualizations results, the model cannot be explained. Hence, this study uses the SHAP visualizations to explain the random forest model educated and tested within this study. to clarify the random forest model educated and tested in this analysis. 4.7. SHAP Visualizations four.7. SHAP Visualizations This research utilizes the SHAP python.