Penerapan Algoritma Optimasi Chaos pada Jaringan Ridge Polynomial untuk Prediksi Jumlah Pengangguran
Rina Pramitasari(1*), Retantyo Wardoyo(2)
(1) 
(2) Departemen Ilmu Komputer dan Elektronika, FMIPA UGM, Yogyakarta
(*) Corresponding Author
Abstract
Abstrak
Ridge polynomial neural network (RPNN) awalnya diusulkan oleh Shin dan Ghosh, dibangun dari jumlah peningkatan order pi-sigma neuron (PSN). RPNN mempertahankan pembelajaran cepat, pemetaan yang kuat dari layer tunggal higher order neural network (HONN) dan menghindari banyaknya bobot karena meningkatnya sejumlah input. Algoritma optimasi chaos digunakan dengan memanfaatkan persamaan logistik yang sensitif terhadap kondisi awal, sehingga pergerakan chaos dapat berubah di setiap keadaan dalam skala tertentu menurut keteraturan, ergodik dan mempertahankan keragaman solusi.
Algoritma Optimasi Chaos diterapkan pada RPNN dan digunakan untuk prediksi jumlah pengangguran di Kalimantan Barat. Proses pelatihan jaringan menggunakan ridge polynomial neural network, sedangkan pencarian nilai awal bobot dan bias jaringan menggunakan algoritma optimasi chaos. Struktur yang digunakan terdiri dari 6 neuron layer input dan 1 neuron layer output. Data diperoleh dari Badan Pusat Statistik.
Hasil dari penelitian ini menunjukkan bahwa algoritma yang diusulkan dapat digunakan untuk prediksi.
Kata kunci—prediksi jumlah pengangguran, jaringan syaraf tiruan, algoritma optimasi chaos, ridge polynomial neural network
Abstract
Ridge polynomial neural network was initially proposed by Shin and Ghosh, made of total increased pi-sigma neural (PSN) orders. Ridge polynomial neural network maintains quick learning, strong mapping of single layer of higher order neural network (HONN) and avoids many weights because total increased inputs. Chaos optimization algorithm is used by utilizing sensitive logistic equation to initial condition, so that chaos movement can change in each condition in specific scale according to orderliness, ergodic, and maintaining solution variety.
Chaos optimization algorithm is applied to ridge polynomial neural network and used to predict total unemployed persons in West Kalimantan. Network training process used ridge polynomial neural network; while, initial values and weights and bias of network were found using Chaos optimization algorithm. Structure used consisted of 6 input layer neurons and one output layer neuron. Data were obtained from Central Statistic Agency.
The results of research indicated that algorithm proposed could be used to predict
Keywords— predict the number of unemployed, neural networks, chaos optimization algorithm, ridge polynomial neural network
Keywords
Full Text:
PDFReferences
[1]Hacib, T., Bihan, Y. L., Mekideche, M. R., Ferkha, N., 2010, Ridge Polynomial Neural Network for Non-destructive Eddy Current Evaluation, Computational Methods for the Innovative Design of Electrical Devices, (Studies in Computational Intelligence 327), Springer, Verlag Berlin Heidelberg.
[2]Zhang, R., Zheng, X., 2010, A New Flatness Pattern Recognition Model Based on Variable Metric Chaos Optimization Neural Network, R. Zhu et al. (Eds.): ICICA 2010, LNCS 6377, pp. 357-364, Springer-Verlag Berlin Heidelberg 2010.
[3]Karnavas, Y. L., Papadopoulos, D. P., 2004, Excitation Control of a Synchronous Machine Using Polynomial Neural Networks, Journal of ELECTRICAL ENGINEERING, VOL. 55, NO. 7-8. 2004, 169-179, ISSN 1335-3632.
[4]Velasquez, J. D., 2010, An Enhanced Hybrid Chaotic Algorithm using Cyclic Coordinate Search and Gradient Techniques,revista de ingenieria. Universidad de los Andes. Bogota, Colombia. rev.ing. ISSN. 0121-4993. Julio – Diciembre de 2010, pp.45-53.
[5]Khoa T. Q. D. dan Nakagawa M., 2007, Neural Network Learning based on Chaos, International Journal of Computer and Information Engineering 1:2 2007
[6]Samarasinghe, S., 2007, Neural Networks for Applied Sciences and Engineering, From Fundamentals to Complex Pattern Recognition, Auerbach Publicationss, New York.
[7]Epitropakis, M. G., Vrahatis, M. N., 2005, Root finding and approximation approaches through neural networks, ACM SIGSAM Bulletin, Vol 39, No. 4, Desember 2005.
[8]Gupta, M. M., Jin, L., Homma, N., 2003, Static and Dynamic Neural Networks, From Fundamentals to Advanced Theory, John Wiley & Sons, Inc., Hoboken, New Jersey
DOI: https://doi.org/10.22146/ijccs.2151
Article Metrics
Abstract views : 2413 | views : 1949Refbacks
- There are currently no refbacks.
Copyright (c) 2013 IJCCS - Indonesian Journal of Computing and Cybernetics Systems
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
View My Stats1