Implementation of Artifical Neural Networks with Multilayer Perceptron for Analysis of Acceptance of Permanent Lecturers

Hartono Hartono (Universitas Potensi Utama)
Muhammad Sadikin (Universitas Potensi Utama)
Dian Maya Sari (Universitas Potensi Utama)
Nur Anzelina (Universitas Potensi Utama)
Silvia Lestari (Universitas Potensi Utama)
Wulan Dari (Universitas Potensi Utama)

Jurnal Mantik
Vol. 4 No. 2 (2020): Augustus: Manajemen, Teknologi Informatika dan Komunikasi (Mantik)

Lecturer acceptance selection is the first step in building an education. The Multilayer Perceptron method can be applied in the case of permanent lecturer admissions. The problem faced in the admission of permanent lecturers is that reception is still subjective. This research will prove the ability of the Multilayer Perceptron algorithm to classify eligibility as a lecturer or not. Inputs from this study were prospective applicants' data, namely age, grade point average (GPA), written test score, interview value, and home base status. Sample data amounted to 100 data. 75% of the data is used as training data, and 25% as test data. The test results of the accuracy of the data are known that the multilayer perceptron neural network method has an accuracy rate of 98.7% and with a ROC Area value of 0.989. This proves that the application of the model used belongs to the classification category very well because it has a ROC value between 0.90-1.00.