Penerapan AlgoritmaSupport Vector MachineUntuk Model Prediksi Kelulusan MahasiswaTepat Waktu

Penerapan AlgoritmaSupport Vector MachineUntuk Model Prediksi Kelulusan MahasiswaTepat Waktu

SVM, graduation, students, algorithm, machine learning

A University can have many student data in their databasebecause many students did not graduate on time.Data mining technique can be used to process student data to predict student graduation on time. Support Vector Machine (SVM) algorithm is one of data mining techniques.Objectives of this research was implementation of SVM algorithm to model the prediction ofstudent graduation on timein private university in Indonesia. This research was conducted using CRISP-DM(Cross Industry Standard Process for Data Mining)method. There arefive stepsin that methodsuch as understanding businessto predict student graduation in time which is not available, data understandingby choosing the right attribute for the next step, data preparationincludes cleaningthe null dataand transforming datainto category which has been specified, modeling was used by implementingdatatraining and data testingon SVM algorithm and evaluationto validate and measure the accuracy of the model. The result of this research shown thataccuracy value of data testing was 94,4% using 90% data training and 10% data testing. This concluded SVM algorithm can be used to model the prediction of student graduation on time

EmyHaryatmi1, Sheila Pramita Hervianti2

https://jurnal.iaii.or.id/index.php/RESTI/issue/view/22

Universitas Gunadarma

30 april 2021

Fajar Bagus W

PDF

Indonesia

Text