ANALISIS ALGORITMA C 4.5, NAÏVE BAYES DAN K-NEAREST NEIGHBOR UNTUK MENENTUKAN PENERIMAAN BEASISWA

Abstract

Scholarships are one of the solutions to overcome the problem of costs for those who are less fortunate. Granting scholarships at Indra Bangsa Vocational School sometimes only focuses on orphaned students and scholarships that have good grades so they don't get scholarships, so this study aims to determine the classification of scholarship recipients at Indra Bangsa Vocational School, for the data used as research units are SMK students from class 10-12 with a total of 56 students, of which 30% will be taken as data testing. With this data mining algorithm, the data attributes used for processing include data on family status, number of dependents of parents, student rankings and parents' income. And based on the data mining process it was found that the Naïve Bayes and K-Nearest Neighbor algorithms are more accurate in eliminating the C4.5 algorithm in classifying scholarship recipients. And based on the results of scholarship testing with the rapid miner application, the accuracy value is obtained that the KNN algorithm gets a higher accuracy value of 88.24% while the C4.5 and Naïve Bayes algorithms get an accuracy value of 82.35%. In this case, it can be seen that the best method is the K-Nearest Neighbor algorithm, then the C4.5 and Naïve Bayes algorithms. The suggestion for this research is to use more attributes and use other algorithms for the testing process. The update from the research made is to use 3 algorithms in the data mining process carried out at the Indra Bangsa Vocational School.