Pemetaan Promosi dalam Penjaringan Calon Mahasiswa Menggunakan Algoritma Backpropagation
Keywords: Backpropagation Algorithm, Forecasting, Students, Promotion Costs, Districts
Promotion requires a large fee if it is not targeted when doing it. Backpropagation is an excellent method of dealing with the problem of recognizing complex patterns. Backprogation neural network each unit in the input layer is connected to each unit in the hidden layer. Student data from 2014 to 2018 is a comparison point. The results of testing of this method are calculations using a sample value of 5 years before using a comparative value of 2014 to 2018 totaling 602 data. This research uses 5-5-1 architecture, epoch 2000 and learning rate so that the data accuracy reaches 71% with an error value of 0.0099. The results of this study are 16 districts that become promotion recommendations. Ordering of forecasting the highest number of students to the smallest number of students, so it can be concluded that this method is very useful in mapping promotions.
Download data is not yet available.
How to Cite
Kurniawan, M. H., Defit, S., & Yunus, Y. (2020). Pemetaan Promosi dalam Penjaringan Calon Mahasiswa Menggunakan Algoritma Backpropagation . Jurnal Informatika Ekonomi Bisnis, 2(1), 21-26. https://doi.org/10.37034/infeb.v2i1.17