HOS-Miner: a system for detecting outlying subspaces of high-dimensional data

Zhang, Ji and Lou, Meng and Ling, Tok Wang and Wang, Hai (2004) HOS-Miner: a system for detecting outlying subspaces of high-dimensional data. In: 30th International Conference on Very Large Data Bases (VLDB'04), 31 Aug - 3 Sept 2004, Toronto, Canada.

[img]
Preview
PDF (Accepted Version)
Zhang_Lou_Ling_Wang_VLDB'04_AV.pdf

Download (111Kb)

Abstract

[Abstract]: We identify a new and interesting high-dimensional outlier detection problem in this paper that is, detecting the subspaces in which given data points are outliers. We call the subspaces in which a data point is an outlier as its Outlying Subspaces. In this paper, we will propose the prototype of a dynamic subspace search system, called HOS-Miner (HOS stands for High-dimensional Outlying Subspaces) that utilizes a sample-based learning process to effectively identify the outlying subspaces of a given point.


Statistics for USQ ePrint 5654
Statistics for this ePrint Item
Item Type: Conference or Workshop Item (Commonwealth Reporting Category E) (Paper)
Refereed: Yes
Item Status: Live Archive
Additional Information: Author's version deposited in accordance with the copyright policy of the publisher.
Depositing User: Dr Ji Zhang
Faculty / Department / School: Historic - Faculty of Sciences - Department of Maths and Computing
Date Deposited: 28 Sep 2009 06:24
Last Modified: 02 Jul 2013 23:23
Uncontrolled Keywords: HOS-Miner; High-dimensional Outlying Subspaces
Fields of Research (FOR2008): 08 Information and Computing Sciences > 0801 Artificial Intelligence and Image Processing > 080109 Pattern Recognition and Data Mining
URI: http://eprints.usq.edu.au/id/eprint/5654

Actions (login required)

View Item Archive Repository Staff Only