Penerapan Data Mining dalam Sistem Pendukung Keputusan Penerimaan Beasiswa KIP-Kuliah Menggunakan Algoritma Naive Bayes

Abstract

Improving access to higher education through the KIP-Kuliah scholarship program faces challenges in the selection process, particularly in determining recipients efficiently and objectively amid evolving policies and socioeconomic conditions. At UIN Alauddin Makassar, the selection process for 2,380 applicants in 2024 resulted in a heavy administrative burden on the Student Affairs Bureau, with final decisions made at the rectorate level. This study aims to develop a decision support system based on the Naive Bayes algorithm to assist in the computerized classification of scholarship eligibility. The model was trained using 350 data entries and tested on 150 entries, achieving an accuracy rate of 96.67%, indicating strong classification performance. The confusion matrix showed 145 correct predictions out of 150 test data, with precision, recall, and F1-score values all above 0.95. The system facilitates automated data management and eligibility prediction, reducing process complexity and accelerating decision-making. Black-box testing confirmed that the system functions according to specifications, suggesting that the implementation of the Naive Bayes algorithm can enhance the scholarship selection process to be more accurate, efficient, and fair—particularly for students from underprivileged backgrounds.