Access Restriction

Author Jensen, R. ♦ Qiang Shen
Source IEEE Xplore Digital Library
Content type Text
Publisher Institute of Electrical and Electronics Engineers, Inc. (IEEE)
File Format PDF
Copyright Year ©2007
Language English
Subject Domain (in DDC) Computer science, information & general works ♦ Special computer methods
Subject Keyword Set theory ♦ Computational intelligence ♦ Glass ♦ Pattern recognition ♦ Rough sets ♦ Forensics ♦ Text processing ♦ Computer science ♦ Discrete transforms ♦ Decision making
Abstract One of the main obstacles facing the application of computational intelligence technologies in pattern recognition (and indeed in many other tasks) is that of dataset dimensionality. To enable pattern classifiers to be effective, a dimensionality minimization step is usually carried out beforehand. Rough set theory has been successfully applied for this as it requires only the supplied data and no other information; most other methods require supplementary knowledge. However, the main limitation of traditional rough set-based selection in the literature is the restrictive requirement that all data is discrete; it is not possible to consider real-valued or noisy data. This has been tackled previously via the use of discretization methods, but may result in information loss. This paper investigates two approaches based on rough set extensions, namely fuzzy-rough and tolerance rough sets, that address these problems and retain dataset semantics. The methods are compared experimentally and utilized for the task of forensic glass fragment identification.
Description Author affiliation: Univ. of Wales, Aberystwyth (Jensen, R.; Qiang Shen)
ISBN 1424412099
ISSN 10987584
Educational Role Student ♦ Teacher
Age Range above 22 year
Educational Use Research ♦ Reading
Education Level UG and PG
Learning Resource Type Article
Publisher Date 2007-07-23
Publisher Place UK
Rights Holder Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Size (in Bytes) 464.64 kB
Page Count 6
Starting Page 1
Ending Page 6

Source: IEEE Xplore Digital Library