부산대학교 도서관

There will be a lot of student-related data produced by a digital campus. The key to raising students' management skills is learning how to analyze and apply these data. In addition to helping schools improve school safety and provide early warning of potentially dangerous situations, the analysis of student behavior data can also be used to explain student behavior using real data, providing quantitative data support for scholarship & grant evaluation. In an effort to improve teaching and learning, predicting student performance within educational settings is a hotly debated topic among researchers. Educators and teachers could create appropriate teaching materials to assist students in studying in accordance with projected outcomes with the aid of effective prediction approaches and characteristics. Using algorithms for machine learning such as SVM, DT, Ensemble, & KNN, together with metrics for precision, accuracy, recall, or F1 score, the purpose of this study is to make a prediction about the performance of students. Support Vector Machines (SVM) obtained the value of 0.9987, 0.9948, 0.9931, and 0.9743, respectively, when compared to Ensembles, Decision Tree (DT), & KNN, for Accuracy, Precision, and Recall, and F1 score.