Claim Missing Document
Check
Articles

Found 1 Documents
Search

Peramalan Harga Pasar Telur Ayam Ras di Kota Malang dengan Menggunakan Metode "PSO-NN" Nadya Sylviani; Arief Andy Soebroto
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 12 (2018): Desember 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (933.805 KB)

Abstract

The increase in population and growth in the farming field was one of the reason of rising in consumption of animal protein. One of the source of animal protein that has been the most preferable for its affordable price and easy to get, is purebred chicken egg. However, the issue in purebred chicken egg sales that has been faced by the market is its fluctuate price. Therefore, the researcher build a forecasting system using Particle Swarm Optimization-Neural Network (PSO-NN). This method proved to be better than the NN training using Backpropagation. PSO training will be carried out to the maximum iteration and the results of the training in the form of optimal weight values will be used as initialization weights for NN training. The type of NN used is Feedforward Neural Network (FFNN). The NN training is done by using 4 lag time, 2 hidden units, and 1 output which represents the result of price forecasting. Based on the result of this research, PSO proved to be able to speed up NN on finding the optimal solution. The lowest MAPE found using PSO-NN in this research is 1.01552%.