Scholarships are a form of assistance given to students who have good academic abilities or to their parents who cannot afford to pay for their education. In its application, the selection process for prospective scholarship recipients is carried out manually by comparing the applicant files. This form of selection is very time-consuming and therefore ineffective and inefficient. Another obstacle is that the decision maker has difficulty in deciding who will get the scholarship, and there are many other obstacles faced. The purpose of this study is to assist decision-makers in the selection process for prospective scholarship recipients easily and quickly. The dataset in this study uses eight attributes with 100 instances. The classification method used is the Naïve Bayes Algorithm. The validation process uses the split test technique. This method is compared with other methods such as C4.5 and KNN. The results of this research process, the proposed method can predict prospective scholarship recipients correctly as evidenced by an accuracy value of 90%. However, other algorithms also get the same accuracy value in the 80:20 split test, the accuracy obtained is different when the 70:30 split test score with nave Baye is ranked first. Seeing the results obtained, the Naïve Bayes algorithm can be applied to a decision support system to determine prospective scholarship recipients properly.
Copyrights © 2021