Prediksi Tingkat Kelulusan Nilai Mahasiswa Terhadap Matakuliah Web Programming Menggunkan Metode Neural Network

Abstract

The application of information technology in the world of education can also produce abundant data about student data and the value of learning produced. Like the leading course scores which greatly affect the number of GPA and graduation because if one of the leading courses such as Web Programming gets a D grade then the student cannot register the Final Project or Thesis. One way is to classify student value data to find out what is the most crucial pre-eminent courses from the first semester, using the Neural Network classification method that has advantages in large data and provides accurate predictions of passing student grades in order to get precise accuracy to classify graduation of excellent subjects in the Information Systems study program. From the results of this study obtained an accuracy value of 95.67% and a ROC graph with an AUC (Area Under Curve) value of 0.939.