Employing Supervised Learning Techniques for College Major Prediction: Empowering Decision-Making in University Admission Systems

Abstract

Introduction: Choosing an appropriate study major is a significant challenge for prospective university students. Many students enroll in programs misaligned with their interests or abilities, leading to major changes, extended study duration, financial losses, and the displacement of other potential students. Addressing this issue requires a systematic approach to assist students in making informed decisions.

Objectives: This study aims to develop an intelligent decision support model to predict the most suitable undergraduate major based on students' academic performance, interpersonal influences, and external factors. In additiona, it seeks to identify the key variables that influence major selection.

Methods: A dataset was collected from 430 participants through a structured questionnaire comprising 18 questions. Various supervised learning techniques, including Support Vector Machine (SVM), Random Forest, and Naïve Bayes, were employed to predict the most suitable major. Furthermore, Random Forest and XGBoost were used to analyze relationships between study majors and input variables to determine the most influential factors.

Results: The models demonstrated that SVM, Random Forest, and Naïve Bayes provided the most accurate predictions. The analysis identified creativity, technical skills, and GPA as the top three factors influencing students' study major selection.

Conclusions: The findings emphasize the importance of data-driven decision-making in academic advising. By leveraging machine learning techniques, institutions and career counselors can guide students toward suitable majors, reducing the likelihood of major changes and optimizing educational resources.