<![CDATA[Graduation accuracy is one of the indicators used in assessing the suitability of undergraduate programs as a functional unit of higher education. Knowing the factors that influence student graduation time helps study programs and faculties make decisions to increase the number of students who graduate on time. The purpose of this research is to obtain an overview of the factors that influence students in the accuracy of completing the study time by using the Machine Learning algorithm, namely Decision Tree, which is expected to have high classification efficiency and good description so that it can increase the number of students graduating on time. The methods used to determine student dropout factors are Classification and Regression Tree (CART) and LightGBM. The data used is the data of undergraduate students of Universitas Muhammadiyah Sumatera Utara in 2019. The quality of classification can be read from the accuracy, sensitivity and specification values. The result using CART is 95.1% with the most influencing factors are GPA, faculty, lecture time and predicate while Lightgbm is 83% with the most influencing factors are GPA, gender, lecture time and faculty. Decision tree can be used to determine student dropout factors because of its high accuracy with GPA being the main factor.]]>
Copyrights © 2024
