FRESHMEN STUDENTS’ ENROLLMENT FORECASTINGSYSTEM WITH DECISION SUPPORT

ABSTRACT

This study aimed to develop a Freshmen Students’ Enrollment Forecasting System with Decision Support for Guangzhou Institute of Science and Technology by applying a feed-forward artificial neural network to predict consultees’ enrollment status. A multi-layer feed-forward network with backpropagation learning was implemented and trained, tested, and validated using 21,690 student records from academic years 2021 to 2024, compiled from admissions records and institutional research databases. The model utilized seven input variables: Applied Program, Graduated Students’ Average Employment Rate of the Program (SER), Total Enrolment Fee-to-Salary Ratio (TSR), Program Rating (PR), whether the consultee’s NCEE score was greater than or equal to the least admission score (IsAbove), Total Enrolment Fee-to-Average Monthly Income Ratio of the consultee’s city (TIR), and willingness to accept program adjustment if the applied program was not met (AcpAdj). These variables were used to classify students as either likely to enroll or not enroll. A random search tuner was employed to identify the optimal hyperparameter configuration, and results showed that a multi-layer neural network with 64 hidden neurons, a learning rate of 0.01, a momentum value of 0.87, and a ReLU activation function achieved the best performance, with a classification accuracy exceeding 93%. The findings demonstrate that the proposed model can effectively support enrolment prediction and assist the institution in optimizing human resource planning and campus facility utilization before the start of a new school year.

Keywords: Artificial neural network, decision support system, enrollment forecasting, feed-forward neural network, higher education admissions
https://doi.org/10.57180/vgad8957