Classification Of Prospective Scholarship Recipients Kartu Indonesia Pintar (KIP) With Decision Tree Algorithm And Naïve Bayes

Abstract

The purpose of this study is to produce a model that can assist in determining prospective new students of STMIK AKBA who receive scholarships. The algorithm used is decision tree and nave Bayes to classify the graduation of prospective recipients of the Indonesian Smart Card (KIP) scholarship. Based on the results of the classification of the decision tree algorithm with the confusion matrix, the accuracy value is 44.12 % and the F1-Score is 34.48 %. If you use the Naive Bayes algorithm, it produces an accuracy value of 76.47 % using data on diploma scores and average report cards. Furthermore, for accuracy without using diploma value data and the average report card is 79.41 %. The results of this study show that nave Bayes has a better performance even though it does not use diploma scores and average report cards. Measurement of the results of the nave Bayes classification with the confusion matrix showed low sensitivity with values ​​of 66.67 % and 58.33 % for the first and second scenarios. Based on the evaluation results, the Naïve Bayes algorithm has a better performance than the Decision Tree algorithm in classifying KIP scholarship recipients.