Predicting School Dropout of International Students in Turkey

This master's thesis research addresses the critical challenge of international student dropout in Turkish higher education institutions. Our study specifically focuses on Yemeni students while developing a methodology that can be adapted for other international student populations.

Research Significance:

International students face unique challenges including language barriers, cultural adaptation difficulties, financial constraints, and limited social support networks. These factors contribute to higher dropout rates compared to domestic students. Our research aims to bridge this gap through data-driven predictive analytics.

Methodology & Innovation:

We employ advanced machine learning techniques including Decision Trees, Random Forest, Artificial Neural Networks, Logistic Regression, and Support Vector Machines to analyze comprehensive datasets. Our model is trained on Yemeni student data but designed with adaptable features for broader international student populations.

Impact & Applications:

The predictive system enables universities to identify at-risk students early in their academic journey, allowing for timely interventions such as academic support, counseling services, financial assistance, and cultural integration programs.