Penerapan Data Mining Klasifikasi C4.5 Pada Penerima Beasiswa di SMK Swasta Anak Bangsa

Millah Sari (STIKOM Tunas Bangsa, Pematangsiantar)
Agus Perdana Windarto (STIKOM Tunas Bangsa, Pematangsiantar)
Harly Okprana (STIKOM Tunas Bangsa, Pematangsiantar)

BEES: Bulletin of Electrical and Electronics Engineering
Vol 1 No 3 (2021): Maret 2021

Data Mining is a series of processes to explore added value in the form of knowledge that has not been known manually from a data set. There are 5 (five) attributes used in this study, namely: Value, Attendance, Semester, Parents' Income (PO), and Number of Dependent Parents (JTO). Based on data processing using Rapid Miner 5.3.0.0 software, an accuracy value of 92.70% is obtained, meaning that the resulting rule is close to 100% correct. Where the results of the feasible precision label class are 92.05% and the inappropriate label is 93.24%. In accordance with these provisions, the results of manual calculations by Rapid Miner testing produce 9 models of rules or rules for Scholarship Recipients. This means that the results of the process carried out by researchers on the calculation of the C4.5 Algorithm and Rapidminer obtained the same and appropriate results. So that testing with Rapid Miner can be said to be successful and can find a decision tree in the case of Scholarship recipients.