Kesan Pernormalan Data Dan Struktur Rangkaian Terhadap Prestasi Rangkaian Neural Dalam Peramalan.(Kajian Kes Kemasukan Pelancong Ke Langkawi)

ABU BAKAR, NAJMI WAHIDI (2002) Kesan Pernormalan Data Dan Struktur Rangkaian Terhadap Prestasi Rangkaian Neural Dalam Peramalan.(Kajian Kes Kemasukan Pelancong Ke Langkawi). Other thesis, Universiti Teknologi Malaysia.

[img]
Preview
PDF
najmi_wahidibin_abu_bakar.pdf

Download (2099Kb) | Preview

Abstract

The application of neural network with back propagation algorithm in field of forecasting is now widely used. However,there are few factors that must be looked into in order to produce a network that will converge and generalize. Basically the learning process of a network is influenced by value of momentum constant,learning rate parameter,number of hidden nodes and selected transfer function. Inappropriate choose of value will result in network not reaching the targeted value. Effects of data transformation and stationary data on the ability of the network to converge and generalize also will be studied. The arrival of tourist to Langkawi by ferry, cruise and flight are used as a case study for this project. Neural network models are built using Matlab 6.1 and Statistica is used to make data stationary.

Item Type: Thesis (Other)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Faculty of Computer Science and Information System > Industrial Computing and Modelling
Depositing User: Unnamed user with email knizam@utm.my
Date Deposited: 04 Jul 2013 07:13
Last Modified: 04 Jul 2013 07:13
URI: http://ir.fsksm.utm.my/id/eprint/1919

Actions (login required)

View Item View Item