Machine Learning Estimation for Course Enrollment using Logistic Regression

Machine Learning is being utilized by educational institutions to effectively manage different processes and transactions. Several colleges and departments are using predicting models to obtain information in the available data that they have. In this study, machine learning models were used to predict the total number of sections or enrollment headcount. The Logistic regression technique was used to identify the attribute that affects the possibility of students enrolling in a specific program. Thus, the results of logistic regression were used to identify the best model in predicting the number of sections or courses to be offered by the programs. The accuracy of Support Vector Machine, K-Nearest Neighbor, Decision Tree, and Neural Network is calculated based on the precision, recall values of the classification matrix. The experimental results show the model can identify the attributes contributing to the choice of the student whether to enroll in a certain program. Moreover, based on the comparison of different models Support Vector Machine got the highest True Positive Rate which is 80% on the ICET program and 70% and higher accuracy in the BSID, BSIC, BSECE, and BS Che.