ANALISIS PERFORMA ALGORITME C4.5 PADA DATASET PENERIMAAN BEASISWA DI STMIK AMIKOM PURWOKERTO

STMIK Amikom Purwokerto is a college that began holding scholarships since 2013. One of the scholarships held at STMIK Amikom Purwokerto is the PPA scholarship. PPA Scholarship (Improved Academic Achievement) is a grant given by the government to outstanding students. The implementation of the scholarship is done by collecting the registration documents at the STMIK Amikom Purwokerto student affairs section. Furthermore, the data was selected using the help of Microsoft Excel, but after the selection process the data must be carefully chosen again manually because the scholarship acceptance criteria are many and the time given by the government is only 1 week, while the files collected by each student are not small. Because the selection must be done again manually, the authors try to analyze the scholarship receipt data with data mining techniques C4.5 using Weka software to produce the most influential attributes and provide useful information for the next scholarship acceptance selection process. The purpose of this study is to determine the performance of the C4.5 algorithm and find out the most influential attributes so that it can be considered for the next scholarship selection process. The method used in this research is C4.5 data mining algorithm method. The results obtained from the study using the C4.5 algorithm obtained accuracy with the imputation of missing values of 63.44% and without the imputation of missing values of 68.45% and accuracy by removing attributes that had no effect of 67.38%. Judging from the three results, the C4.5 algorithm has a pretty good accuracy. The most influential attributes are achievement, semester, age, activeness, getting scholarship and GPA.