Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/14519
Title: Performance enhancement of data-driven models with derived quality data and data assimilation in water resources
Authors: DOAN CHI DUNG
Keywords: Data Reduction, Data Cleaning, Clustering, Evolutionary Algorithms, Data Assimilation, Data-Driven Model
Issue Date: 1-Feb-2005
Source: DOAN CHI DUNG (2005-02-01). Performance enhancement of data-driven models with derived quality data and data assimilation in water resources. ScholarBank@NUS Repository.
Abstract: Real data are often contaminated with many sources of noise. They may also suffer from the problem of inconsistency, redundancy, etc. The study proposed two schemes to deal with real data so that good performance of data-driven models can be ascertained: (1) a scheme to enhance data quality; and (2) a data assimilation technique for noisy data. With data quality enhancement scheme, a highly compact, effective and efficient input data set is derived. It results in a forecasting model, trained with this reduced data, which yields equally high prediction accuracy as that resulting from forecasting model trained with the complete data set. With data assimilation, results showed that noisy data was dealt with effectively. The proposed data assimilation schemes are of particular advantage when the system under consideration has no information on its functional derivatives. Results also revealed that the performance of neural network can be further improved particularly when the data considered are of relatively high or very high noise levels.
URI: http://scholarbank.nus.edu.sg/handle/10635/14519
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
!01 Page Cover.pdf5.8 kBAdobe PDF

OPEN

NoneView/Download
!03 Page Title.pdf14.86 kBAdobe PDF

OPEN

NoneView/Download
!04 Acknowledgements.pdf9.63 kBAdobe PDF

OPEN

NoneView/Download
!05 Table of Contents.pdf14.82 kBAdobe PDF

OPEN

NoneView/Download
!06 Summary.pdf10.85 kBAdobe PDF

OPEN

NoneView/Download
!07 Nomenclature, List of Figures & Tables.pdf83.89 kBAdobe PDF

OPEN

NoneView/Download
01 Chapter 1.pdf38.93 kBAdobe PDF

OPEN

NoneView/Download
02 Chapter 2.pdf75.15 kBAdobe PDF

OPEN

NoneView/Download
03 Chapter 3.pdf322.48 kBAdobe PDF

OPEN

NoneView/Download
04 Chapter 4.pdf219.15 kBAdobe PDF

OPEN

NoneView/Download
05 Chapter 5.pdf221.97 kBAdobe PDF

OPEN

NoneView/Download
06 Chapter 6.pdf16.94 kBAdobe PDF

OPEN

NoneView/Download
07 References.pdf32.44 kBAdobe PDF

OPEN

NoneView/Download
08 List of Publications.pdf12.06 kBAdobe PDF

OPEN

NoneView/Download

Page view(s)

223
checked on Dec 2, 2017

Download(s)

2,137
checked on Dec 2, 2017

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.