Relevant attribute discovery in high dimensional data based on rough sets and unsupervised classification: application to Leukemia gene expressions

From National Research Council Canada

Alternative titleRelevant attribute discovery in high dimensional data based on rough sets applications to Leukemia gene expressions
Download
  1. (PDF, 280 KB)
DOIResolve DOI: https://doi.org/10.1007/11548706_38
AuthorSearch for: 1; Search for: 1
Name affiliation
  1. National Research Council of Canada. NRC Institute for Information Technology
FormatText, Book chapter
Proceedings titleRough Sets, Fuzzy Sets, Data Mining, and Granular Computing
Series titleLecture Notes in Computer Science; Volume 3642
Lecture Notes in Artificial Intelligence; Volume 3642
ConferenceThe Tenth International Conference on Rough Sets, Fuzzy Sets, Data Mining and Granular Computing (RSFDGrC 2005), August 31 - September 3, 2005, Regina, Saskatchewan, Canada
ISSN0302-9743
1611-3349
ISBN978-3-540-28660-8
978-3-540-31824-8
Pages362–371
Subjectacute myeloid leukemia; acute lymphoblastic leukemia; high dimensional data; decision attribute; remote host
Abstract
A pipelined approach using two clustering algorithms in combination with Rough Sets is investigated for the purpose discovering important combination of attributes in high dimensional data. In many domains, the data objects are described in terms of a large number of features, like in gene expression experiments, or in samples characterized by spectral information. The Leader and several k-means algorithms are used as fast procedures for attribute set simplification of the information systems presented to the rough sets algorithms. The data submatrices described in terms of these features are then discretized w.r.t the decision attribute according to different rough set based schemes. From them, the reducts and their derived rules are extracted, which are applied to test data in order to evaluate the resulting classification accuracy. An exploration of this approach (using Leukemia gene expression data) was conducted in a series of experiments within a high-throughput distributed-computing environment. They led to subsets of genes with high discrimination power. Good results were obtained with no preprocessing applied to the data.
PublisherSpringer
LanguageEnglish
Peer reviewedYes
NRC numberNRCC 48122
NPARC number8913287
Export citationExport as RIS
Report a correctionReport a correction
Record identifiered378d76-e02c-49d9-9c9a-4b18c393e1d6
Record created2009-04-22
Record modified2020-09-15
Report a problem on this page
Date modified: