STUDENT GRADUATION TIME PREDICTION USING LOGISTIC REGRESSION, DECISION TREE, SUPPORT VECTOR MACHINE, AND ADABOOST ENSEMBLE LEARNING

Abstract

Universities in Indonesia are working hard to improve the graduation rates of their students as it is considered a measure of success and quality in terms of accreditation. This study focuses on analyzing the effectiveness of machine learning algorithms, regression, Support Vector Machine (SVM) Decision Tree and ensemble learning, with AdaBoost wether the Computer Science students will graduate on time or not. The data used for this analysis consists of student records from 2015 to 2019. Includes 14 variables. To understand the relationships between these variables a two-dimensional visualization called a Heatmap was employed. The research findings indicate that the Support Vector Machine (SVM) and AdaBoost Decision Tree (DT) algorithm performs better than the other algorithms. The Decision Tree and AdaBoost (DT) model achieved an F1- score of 0,76 and 0,82. This research contributes towards enhancing education management by facilitating decision making to ensure timely graduation, for student