EMAIL: PASSWORD:
Front Office
UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya


Kampus ITS Sukolilo - Surabaya 60111

Phone : 031-5921733 , 5923623
Fax : 031-5937774
E-mail : libits@its.ac.id
Website : http://library.its.ac.id

Support (Customer Service) :
timit_perpus@its.ac.id




Welcome..guys!

Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT


Moh. Fandika Aqsa


Davi Wahyuni


Tondo Indra Nyata


Anis Wulandari


Ansi Aflacha




ITS » Undergraduate Theses » Sistem Informasi - S1
Posted by dewi007 at 13/05/2009 15:24:24  •  5990 Views


METODE PERAMALAN FUZZY GROUP BERBASIS JARINGAN SARAF TIRUAN PADA PREDIKSI NILAI TUKAR MATA UANG

NEURAL-NETWORK-BASED FUZZY GROUP FORECASTING METHOD FOR FOREIGN EXCHANGE RATES PREDICTION

Author :
Darmadi, Teguh 




ABSTRAK

Peramalan nilai tukar mata uang dengan menggunakan Jaringan Saraf Tiruan JST telah digunakan secara luas dan memberikan hasil yang bagus. Hal ini disebabkan kemampuan JST untuk menangkap karakter volatilitas yang tinggi dan nonlinier pada pasar uang. Namun JST merupakan suatu paradigma pembelajaran yang sangat tidak stabil dan terlalu sensitif jika diterapkan pada peramalan nilai tukar mata uang.. Misalnya dengan menggunakan struktur JST yang sama namun berbeda training set maka akan didapat perbedaan hasil yang cukup besar. Hal ini dapat menyebabkan kesulitan dalam menentukan hasil ramalan yang akan digunakan. Untuk menyelesaikan permasalahan tersebut diajukan sebuah pendekatan peramalan baru yang disebut peramalan fuzzy group berbasis jaringan saraf tiruan. Dengan peramalan fuzzy group berbasis jaringan saraf tiruan ini hasil prediksi yang berbeda-beda dari beberapa model JST dapat dijadikan satu hasil yang cukup akurat. Pada metode yang diajukan ini terlebih dahulu digunakan beberapa model JST dimana tiap model JST melakukan beberapa prediksi dengan training set yang berbeda-beda. Kemudian dilakukan fuzzification terhadap hasil prediksi dari setiap model JST sehingga menjadi representasi prediksi fuzzy. Selanjutnya representasi prediksi fuzzy ini diagregasi menjadi satu fuzzy group. Terakhir dilakukan defuzzification terhadap fuzzy group yang diperoleh menjadi satu nilai crisp. Untuk kepentingan ilustrasi dan testing dilakukan beberapa contoh percobaan prediksi terhadap tiga pasang mata uang. Sebagai perbandingan digunakan struktur JST feed-forward untuk melakukan peramalan. Hasil uji coba memperlihatkan bahwa peramalan fuzzy group berbasis JST mampu mengatasi masalah stabilitas dan sensitivitas pada JST. Selain itu keakuratan peramalan juga menjadi lebih baik daripada JST feed-forward. Walaupun begitu peramalan fuzzy group berbasis JST juga memiliki kelemahan yaitu waktu eksekusi yang jauh lebih lama daripada JST feed-forward.


ABSTRACT

Foreign exchange FOREX rates forecasting using Artificial Neural Network ANN has been used widely and obtaining some good results. ANN is used because of its ability to capture the high volatility and nonlinierity character within FOREX market. However ANN is a learning paradigm which is very unstable and too sensitive for foreign exchange rates forecasting. For example using the same ANN structure with different training sets will obtain great difference on results. These characteristic can cause confusion to determine which result will be used. A novel forecasting approach which called neural-network-based fuzzy group forecasting is proposed to solve these problem. With using neural-network-based fuzzy group forecasting the prediction results from some ANN models could be converted to be an adequate accurate result. In the proposed method some ANN models where each model doing some prediction using different training sets are firstly used. Then fuzzification is done toward the single prediction results produced by each ANN model to become fuzzy prediction representations. Subsequently these fuzzy prediction representations are aggregated into a fuzzy group consensus. Finally the fuzzy group that obtained is defuzzified into a crisp value. For illustration and testing purposes some prediction experiments using three FOREX pairs are presented. As a comparison some feed-forward neural networks are used. Experimental results reveal that the neural-network-based fuzzy group forecasting can solve the ANNs stability and sensitivity problem. In the other the prediction accuracy is better than feed-forward neural network. Nevertheless neural-network-based fuzzy group forecasting have a disadvantage that is the running time cost more time than feed-forward neural network.



KeywordsJaringan Saraf Tiruan; Peramalan Fuzzy Group; Prediksi Nilai Tukar Mata Uang
 
Subject:  jaringan saraf tiruan
Contributor
  1. Wiwik Anggraeni, S.Si, M.Kom
    Rully Soelaiman, S.Kom, M.Kom
Date Create: 13/05/2009
Type: Text
Format: pdf.
Language: Indonesian
Identifier: ITS-Undergraduate-3100009034201
Collection ID: 3100009034201
Call Number: RSSI 006.32 Dar m


Source
Undergraduate theses of Information System Engineering, RSSI 006.32 Dar m, 2008

Coverage
ITS Community Only

Rights
Copyright @2008 by ITS Library. This publication is protected by copyright and permission should be obtained from the ITS Library prior to any prohibited reproduction, storage in a retrievel system, or transmission in any form or by any means, electronic, mechanical, photocopying, recording, or likewise. For information regarding permission(s), write to ITS Library




[ Download - Summary ]

ITS-Undergraduate-3100009034201-4119.pdf




 Similar Document...




! ATTENTION !

To facilitate the activation process, please fill out the member application form correctly and completely

Registration activation of our members will process up to max 24 hours (confirm by email). Please wait patiently

POLLING

Bagaimana pendapat Anda tentang layanan repository kami ?

Bagus Sekali
Baik
Biasa
Jelek
Mengecewakan





You are connected from 3.214.224.224
using CCBot/2.0 (https://commoncrawl.org/faq/)



Copyright © ITS Library 2006 - 2020 - All rights reserved.
Dublin Core Metadata Initiative and OpenArchives Compatible
Developed by Hassan