Factors Influencing The Drop-Out Rate Of International Students In South Korea: Application Of Big Data, Social Network, And Machine Learning Analyses

Abstract

This study aims to identify the key factors influencing the attrition rate of international students in South Korea by utilizing big data analysis, machine learning, and social network analysis. By analyzing 643 academic papers related to international students, this research seeks to propose measures to improve their retention rates. Furthermore, machine learning techniques are employed to pinpoint the most significant variables affecting student attrition. The analysis covers 221 four-year universities in Korea, focusing on variables such as the type of institution, student capacity, and geographical location. Results show that these universities can be categorized into seven nodes, with the admission competition rate of new students being the most influential factor; higher competition correlates with lower attrition rates. This study enhances objectivity and validity by employing advanced statistical techniques not used in previous research, providing scientific evidence to support policy improvements.