PERFORMANCE ANALYSIS OF DATA MINING CLASSIFICATION METHODS USING C4.5 ALGORITHM FOR STUDENT GRADUATION PREDICTION (CASE STUDY AT SYEDZA SAINTIKA STIKES)

ABSTRACT The passing rate is one of the parameters of the effectiveness of educational institutions. Decreasing student graduation rate affects higher education accreditation. The college database stores student administrative and academic data, if it is explored properly using data mining techniques, it can be seen patterns or knowledge to make decisions. The C4.5 algorithm is an algorithm used to generate decision trees. This method is popular because it is able to classify as well as show the relationship between attributes. This study used data from students of the 2014 and 2015 class of Public Health study programs. The variables used in this study were: NIM, name, gender, entry status, GPA, area of origin and employment status. Based on the test results by measuring the performance of the method, it is known that C4.5 has a high accuracy value of 97.83%. From the accuracy value, it can be concluded that the C4.5 algorithm has a good performance in predicting the timeliness of student graduation. Key w ords: DataMining, Classification, Naive Bayes, Accuracy of Student Graduation