TY  - EJOU
AU  - Kanwal, Bushra 
AU  - Shoukat, Rana Saud 
AU  - Rehman, Saif Ur 
AU  - Kundi, Mahwish 
AU  - AlSaedi, Tahani 
AU  - Alahmadi, Abdulrahman 

TI  - A New Framework for Scholarship Predictor Using a Machine Learning Approach
T2  - Intelligent Automation \& Soft Computing

PY  - 2024
VL  - 39
IS  - 5
SN  - 2326-005X

AB  - Education is the base of the survival and growth of any state, but due to resource scarcity, students, particularly at the university level, are forced into a difficult situation. Scholarships are the most significant financial aid mechanisms developed to overcome such obstacles and assist the students in continuing with their higher studies. In this study, the convoluted situation of scholarship eligibility criteria, including parental income, responsibilities, and academic achievements, is addressed. In an attempt to maximize the scholarship selection process, numerous machine learning algorithms, including Support Vector Machines, Neural Networks, K-Nearest Neighbors, and the C4.5 algorithm, were applied. The C4.5 algorithm, owing to its efficiency in the prediction of scholarship beneficiaries based on extraneous factors, was capable of predicting a phenomenal 95.62% of predictions using extensive data of a well-esteemed government sector university from Pakistan. This percentage is 4% and 15% better than the remainder of the methods tested, and it depicts the extent of the potential for the technique to enhance the scholarship selection process. The Decision Support Systems (DSS) would not only save the administrative cost but would also create a fair and transparent process in place. In a world where accessibility to education is the key, this research provides data-oriented consolidation to ensure that deserving students are helped and allowed to get the financial assistance that they need to reach higher studies and bridge the gap between the demands of the day and the institutions of intellect.
KW  - Education; data mining; C4.5 algorithm; decision support system; scholarship guarantee; machine learning

DO  - 10.32604/iasc.2024.054645