The Prediction of Late Tuition Fees at Muhammadiyah University of East Kalimantan Using the Logistic Regression Method

Abstract

Muhammadiyah University of East Kalimantan in carrying out its operational activities relies on funds from students, one of which is tuition fees (SPP). This creates problems if students are late in making tuition payments, because tuition payments can improve the quality of education and facilitie. Therefore, the purpose of this study is to determine indicators, then implement the logistic regression algorithm by dividing the data into 70:30 and evaluating the performance results of the algorithm using the confusion matrix. The data obtained are sourced from the Academic Administration Section and the Financial Administration Section as many as 12,408 data with several attributes such as the faculty, study program, class, gender, father's income, mother's income, father's education, mother's education and label (late or not late). From the results of the data test that has been carried out, it gets the results of an accuracy of 55.89%.