An adaptive and dynamic dimensionality reduction method for high-dimensional indexing

Shen, H. T., Zhou, X. and Zhou, A. (2007) An adaptive and dynamic dimensionality reduction method for high-dimensional indexing. The VLDB Journal, 16 2: 219-234. doi:10.1007/s00778-005-0167-3

Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads

Author Shen, H. T.
Zhou, X.
Zhou, A.
Title An adaptive and dynamic dimensionality reduction method for high-dimensional indexing
Journal name The VLDB Journal   Check publisher's open access policy
ISSN 1066-8888
Publication date 2007-04
Year available 2005
Sub-type Article (original research)
DOI 10.1007/s00778-005-0167-3
Volume 16
Issue 2
Start page 219
End page 234
Total pages 16
Place of publication Pacific Grove, Calif., U.S.A.
Publisher Springer
Collection year 2008
Language eng
Subject C1
280108 Database Management
700103 Information processing services
Abstract The notorious "dimensionality curse" is a well-known phenomenon for any multi-dimensional indexes attempting to scale up to high dimensions. One well-known approach to overcome degradation in performance with respect to increasing dimensions is to reduce the dimensionality of the original dataset before constructing the index. However, identifying the correlation among the dimensions and effectively reducing them are challenging tasks. In this paper, we present an adaptive Multi-level Mahalanobis-based Dimensionality Reduction (MMDR) technique for high-dimensional indexing. Our MMDR technique has four notable features compared to existing methods. First, it discovers elliptical clusters for more effective dimensionality reduction by using only the low-dimensional subspaces. Second, data points in the different axis systems are indexed using a single B+-tree. Third, our technique is highly scalable in terms of data size and dimension. Finally, it is also dynamic and adaptive to insertions. An extensive performance study was conducted using both real and synthetic datasets, and the results show that our technique not only achieves higher precision, but also enables queries to be processed efficiently. Copyright Springer-Verlag 2005
Keyword Computer Science, Hardware & Architecture
Computer science, Information systems
High-dimensional indexing
Dimensionality reduction
Correlated clustering
Q-Index Code C1
Q-Index Status Confirmed Code

Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 12 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 18 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Wed, 15 Aug 2007, 07:21:03 EST