Structural Risk Minimisation based gene expression profiling analysis

Xue Wen Chen, Byron Gerlach, Dechang Chen, Zhenqiu Liu

Research output: Contribution to journalArticle

Abstract

For microarray based cancer classification, feature selection is a common method for improving classifier generalisation. Most wrapper methods use cross validation methods to evaluate feature sets. For small sample problems like microarray, however, cross validation methods may overfit the data. In this paper, we propose a Structural Risk Minimisation (SRM) based method for gene selection in cancer classification. SRM principle allows for reducing the probable bound on generalisation error and thus avoids overfitting problems. The experimental results show that the proposed method produces significantly better performance than general wrapper methods that use cross validations.

Original languageEnglish (US)
Pages (from-to)153-169
Number of pages17
JournalInternational Journal of Bioinformatics Research and Applications
Volume3
Issue number2
DOIs
StatePublished - Jul 2 2007

Fingerprint

Gene Expression Profiling
Microarrays
Gene expression
Feature extraction
Classifiers
Genes
Neoplasms

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering
  • Health Informatics
  • Clinical Biochemistry
  • Health Information Management

Cite this

@article{9cde1d5ba120400fa7267598dec012ec,
title = "Structural Risk Minimisation based gene expression profiling analysis",
abstract = "For microarray based cancer classification, feature selection is a common method for improving classifier generalisation. Most wrapper methods use cross validation methods to evaluate feature sets. For small sample problems like microarray, however, cross validation methods may overfit the data. In this paper, we propose a Structural Risk Minimisation (SRM) based method for gene selection in cancer classification. SRM principle allows for reducing the probable bound on generalisation error and thus avoids overfitting problems. The experimental results show that the proposed method produces significantly better performance than general wrapper methods that use cross validations.",
author = "Chen, {Xue Wen} and Byron Gerlach and Dechang Chen and Zhenqiu Liu",
year = "2007",
month = "7",
day = "2",
doi = "10.1504/IJBRA.2007.013600",
language = "English (US)",
volume = "3",
pages = "153--169",
journal = "International Journal of Bioinformatics Research and Applications",
issn = "1744-5485",
publisher = "Inderscience Enterprises Ltd",
number = "2",

}

Structural Risk Minimisation based gene expression profiling analysis. / Chen, Xue Wen; Gerlach, Byron; Chen, Dechang; Liu, Zhenqiu.

In: International Journal of Bioinformatics Research and Applications, Vol. 3, No. 2, 02.07.2007, p. 153-169.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Structural Risk Minimisation based gene expression profiling analysis

AU - Chen, Xue Wen

AU - Gerlach, Byron

AU - Chen, Dechang

AU - Liu, Zhenqiu

PY - 2007/7/2

Y1 - 2007/7/2

N2 - For microarray based cancer classification, feature selection is a common method for improving classifier generalisation. Most wrapper methods use cross validation methods to evaluate feature sets. For small sample problems like microarray, however, cross validation methods may overfit the data. In this paper, we propose a Structural Risk Minimisation (SRM) based method for gene selection in cancer classification. SRM principle allows for reducing the probable bound on generalisation error and thus avoids overfitting problems. The experimental results show that the proposed method produces significantly better performance than general wrapper methods that use cross validations.

AB - For microarray based cancer classification, feature selection is a common method for improving classifier generalisation. Most wrapper methods use cross validation methods to evaluate feature sets. For small sample problems like microarray, however, cross validation methods may overfit the data. In this paper, we propose a Structural Risk Minimisation (SRM) based method for gene selection in cancer classification. SRM principle allows for reducing the probable bound on generalisation error and thus avoids overfitting problems. The experimental results show that the proposed method produces significantly better performance than general wrapper methods that use cross validations.

UR - http://www.scopus.com/inward/record.url?scp=34250886251&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34250886251&partnerID=8YFLogxK

U2 - 10.1504/IJBRA.2007.013600

DO - 10.1504/IJBRA.2007.013600

M3 - Article

C2 - 18048186

AN - SCOPUS:34250886251

VL - 3

SP - 153

EP - 169

JO - International Journal of Bioinformatics Research and Applications

JF - International Journal of Bioinformatics Research and Applications

SN - 1744-5485

IS - 2

ER -