Please use this identifier to cite or link to this item: https://une.intersearch.com.au/unejspui/handle/1959.11/2518
Title: Dimensionality Reduction for Non-vectorial Data
Contributor(s): Guo, Yi (author); Gao, Junbin (supervisor); Kwan, Paul (supervisor)
Conferred Date: 2008
Copyright Date: 2008
Open Access: Yes
Handle Link: https://hdl.handle.net/1959.11/2518
Abstract: Dimensionality Reduction (DR) is an important step in many advanced applications such as exploratory data analysis and manifold learning. Its main goal is to discover the mappings of the input data in a much lower dimensional space or the so-called latent space without incurring unnecessary information loss. In most existing DR algorithms, the main objective is to preserve relational structure among objects of the input space in the latent space by minimizing the inconsistency between two similarity/dissimilarity measures, one for the input data and the other for the embedded data, via a separate matching objective function. Based on this observation, a new dimensionality reduction method called Twin Kernel Embedding (TKE) is proposed. TKE addresses the problem of embedding non-vectorial data that is difficult for conventional methods in practice due to the lack of efficient vectorial representation. TKE solves this problem by minimizing the inconsistency between the similarity measures captured respectively by their kernel Gram matrices in the two spaces. This algorithm is proven to be effective on some real world data sets and has been successfully applied to protein visualization, kernel learning, fingerprint classification etc. TKE is further extended to novel samples by introducing the backward mapping which is incorporated into the objective function as either substitution of all embeddings or regularization terms which generate BCTKE and RCTKE algorithms respectively. Intuitively, the mapping function can be integrated into any other host DR algorithms as a solution to the so-called out-of-sample problem. This thesis starts with the analysis of the existing DR methods. Based on the understanding of their common features, we will show the development of the TKE algorithms and the details on their behaviors at length. We present not only a series of new algorithms, but also the aspects of the design including the origin of the ideas, observations and implementation. This research provides a stepping stone for new algorithmic design in Dimensionality Reduction.
Publication Type: Thesis Doctoral
Rights Statement: Copyright 2008 - Yi Guo
HERDC Category Description: T2 Thesis - Doctorate by Research
Other Links: http://www.scirp.org/Journal/Abstract.aspx?paperID=66&JournalID=30
Statistics to Oct 2018: Visitors: 161
Views: 172
Downloads: 27
Appears in Collections:Thesis Doctoral

Files in This Item:
13 files
File Description SizeFormat 
open/SOURCE05.pdfThesis, part 213.4 MBAdobe PDF
Download Adobe
View/Open
open/SOURCE06.pdfThesis, part 324.29 MBAdobe PDF
Download Adobe
View/Open
open/SOURCE07.pdfThesis, part 419.91 MBAdobe PDF
Download Adobe
View/Open
open/SOURCE04.pdfThesis, part 114.48 MBAdobe PDF
Download Adobe
View/Open
open/SOURCE08.pdfThesis, part 513.04 MBAdobe PDF
Download Adobe
View/Open
open/SOURCE03.pdfAbstract8.18 MBAdobe PDF
Download Adobe
View/Open
open/SOURCE10.pdfThesis, part 77.26 MBAdobe PDF
Download Adobe
View/Open
open/SOURCE09.pdfThesis, part 612.42 MBAdobe PDF
Download Adobe
View/Open
Show full item record

Page view(s)

118
checked on Feb 8, 2019

Download(s)

144
checked on Feb 8, 2019
Google Media

Google ScholarTM

Check

SCOPUSTM   
Citations

 

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