Abstract
The determination of the student's single tuition fee grade at XYZ University still faces problems, one of which is the inconsistency of supporting variables which change each year based on the number of students and the level of student welfare. Determination of the class of student's single tuition fee based on the web that applies the backpropagation algorithm is designed to produce an adaptive decision-making system for changing variable determinants so that every year it is not necessary to adjust the weight in determining the student's single tuition fee group. In building a decision support system to determine the student's single tuition fee group, there are several stages including data collection, data processing, backpropagation implementation, evaluation and presentation of the model, and implementation of a web-based decision support system. During the evaluation and presentation of the model, several experiments were carried out by considering various parameters to get the best accuracy results. The accuracy of the best model produced in this study reached 86% with output in the form of a web-based decision support system that is ready to be managed by the Center for Information and Computer Technology (PUSTIKOM) of XYZ University for use by new students.
Highlights
ABSTRAK Penentuan golongan UKT di Universitas XYZ masih menemui permasalahan, salah satunya adalah inkonsistensi variabel-variabel pendukung yang mana berubah-ubah pada setiap tahunnya berdasarkan jumlah dan kemampuan ekonomi mahasiswa yang diterima
of which is the inconsistency of supporting variables which change each year based on the number of students
Determination of the class of student's single tuition fee based on the web that applies the backpropagation algorithm is designed to produce an
Summary
Di Universitas XYZ, kebijakan penentuan UKT berubah-ubah pada setiap tahunnya berdasarkan jumlah dan kemampuan ekonomi mahasiswa yang diterima yang mana hal tersebut mengacu pada peraturan permenristekdikti no. Maka pada penelitian ini kami akan mengembangkan sistem informasi yang menerapkan Sistem Pengambilan Keputusan dalam penentuan biaya UKT. Model backpropagation memiliki keunggulan tingkat akurasi dibandingkan dengan metode single-layer perceptron awal seperti yang ditunjukan pada penelitian [15] dan [16]. Berdasarkan hal tersebut, maka dalam penelitian ini kami akan menggunakan metode backpropagation dalam pengembangan model Sistem Pengambilan Keputusan untuk menentukan biaya UKT yang selanjutnya diimplementasikan pada Sistem Informasi terkait. Pada penelitian ini proses penghitungan backpropagation dilakukan secara berulang untuk mendapatkan hasil akurasi terbaik dengan mempertimbangkan aspek threshold, learning rate, jumlah hidden layer, dan jumlah node pada hidden layer
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have