Model to Determine the Number of Factors for Neural Network Forecasting System

CHUR > College of Management > Industrial Management > Seminar Papers > Item 987654321/29022

Please use this identifier to cite or link to this item: http://chur.chu.edu.tw/handle/987654321/29022

Title:	Model to Determine the Number of Factors for Neural Network Forecasting System
Authors:	杜瑩美 Tu, Ying Mei
Contributors:	工業管理學系 Industrial Management
Keywords:	類神經網路;迴歸分析;預測模式;資料樣本 Neural network;Linear regression;Forecasting model;Number of factors
Date:	2012
Issue Date:	2014-06-27 00:15:29 (UTC+8)
Abstract:	類神經網路如同大多數其他預測手法，歷史資料樣本的多寡及詳細程度會嚴重影響其預測結果的優劣，若樣本太過精簡則無法表徵母體的行為模式，因此，無法得到滿意的預測品質；反之，過大的樣本雖然可以得到較佳的預測結果，卻會增加資料收集以及模式執行上的困難。因此，如何訂定一個適合的樣本大小及精確度，將會是類神經網路在執行預測的的一個關鍵。本研究以半導體晶圓廠機台群組工件到達量預測問題為例，提出一套以迴歸分析為基礎的最佳輸入因子數量決策模式。利用迴歸分析找出因子數量、預測複雜度指標以及類神經網路預測準確度之間的關係，利用此 Neural network is a good forecasting technique for a complicated system. Nonetheless, the forecast accuracy of neural network is just like other forecasting techniques’ to be seriously impacted by both completeness and details of sample. If the number of
Appears in Collections:	[Industrial Management] Seminar Papers

Files in This Item:

File	Description	Size	Format
s_m515_0465.pdf		29Kb	Adobe PDF	34	View/Open