New Student Admission Forecasting Model with Support Vector Machine Method: Case Study of Bali State Polytechnic

Every educational institution, both formal and non-formal, organizes new student admissions every year. This process requires institutions to improve the quality of education, services, and accreditation, both in terms of student competence, facilities, and infrastructure. Therefore, effective and efficient planning is needed, especially in making strategic decisions. This research aims to forecast the number of new student admissions using the Support Vector Machine (SVM) method. SVM is one of the artificial intelligence techniques known to have a high level of accuracy in data analysis and forecasting. The results showed that the SVM method was able to produce predictions with a low error rate. The test results using Root Mean Square Error (RMSE) show that the Electrical Engineering study program has the best RMSE value of 7.292, making it the study program with the highest level of forecasting accuracy in this study. This finding proves that the SVM method can be effectively implemented in forecasting new student admissions, so that it can help institutions in developing better and data-based admission strategies.