Evaluation of the Effectiveness and Success of the YLSY Scholarship Program With Content Analysis and Machine Learning Methods

The Selection and Placement of Students to be Sent Abroad for Graduate Studies (YLSY) scholarship program is implemented by Türkiye to develop qualified human resources and foster scientific advancement. Based on this objective, the present study aims to evaluate the effectiveness of the program by identifying the factors that influence the academic success of scholarship recipients and examining the results from a multidimensional perspective. To achieve a comprehensive evaluation, An exploratory sequential mixed method was employed. In the qualitative phase, semi-structured interviews were conducted with scholarship recipients, and the data were analyzed through content analysis to determine key needs and challenges. Based on these findings, four success prediction models were developed in the quantitative phase using different machine learning algorithms. The models were designed to predict academic success in terms of indexed article, other article, citation, and project. The indexed article model trained with a boosted decision tree algorithm achieved an accuracy rate of 77.4%, while the other article model trained with a decision jungle algorithm reached 71.1%. The citation model, also using a boosted decision tree, reached 81.6%, and the project model trained with a support vector machine achieved the highest performance with 84.3%. These findings demonstrate the potential of data-driven approaches in forecasting academic success and offer valuable insights for enhancing scholarship programs. Future research should test the generalizability of these models across disciplines and examine their integration into digital decision support systems to improve program impact and efficiency.