Publication detail

Stability of Feature Selection Algorithms and its Influence on Prediction Accuracy in Biomedical Datasets

DROTÁR, P. SMÉKAL, Z.

Original Title

Stability of Feature Selection Algorithms and its Influence on Prediction Accuracy in Biomedical Datasets

English Title

Stability of Feature Selection Algorithms and its Influence on Prediction Accuracy in Biomedical Datasets

Type

conference paper

Language

en

Original Abstract

Feature selection techniques become significant part of many bioinformatics and biomedical applications. Choosing the important features is essential for biomarker discovery, provide better understanding of the data and potentially improve prediction performance. However, as the number of samples in dataset is small, the feature selection tends to be unstable. In this paper, the stability of five popular feature selection techniques is compared and analyzed when original dataset is perturbed by adding, removing or simply resampling the original observations. Next, the feature selection techniques are used as filter prior to AdaBoost classifier and their influence on classification accuracy and Mathews correlation coefficient is compared.

English abstract

Feature selection techniques become significant part of many bioinformatics and biomedical applications. Choosing the important features is essential for biomarker discovery, provide better understanding of the data and potentially improve prediction performance. However, as the number of samples in dataset is small, the feature selection tends to be unstable. In this paper, the stability of five popular feature selection techniques is compared and analyzed when original dataset is perturbed by adding, removing or simply resampling the original observations. Next, the feature selection techniques are used as filter prior to AdaBoost classifier and their influence on classification accuracy and Mathews correlation coefficient is compared.

Keywords

feature selection, stability, Dunne stability index, bioinformatics, Adaboost

Released

27.10.2014

Publisher

IEEE

Location

Bangkok

ISBN

9781479940752

Book

TENCON 2011 - 2011 IEEE Region 10 Conference

Pages from

1

Pages to

4

Pages count

4

URL

Documents

BibTex


@inproceedings{BUT110176,
  author="Peter {Drotár} and Zdeněk {Smékal}",
  title="Stability of Feature Selection Algorithms and its Influence on Prediction Accuracy in Biomedical Datasets",
  annote="Feature selection techniques become significant part
of many bioinformatics and biomedical applications. Choosing
the important features is essential for biomarker discovery,
provide better understanding of the data and potentially improve
prediction performance. However, as the number of samples in
dataset is small, the feature selection tends to be unstable. In this
paper, the stability of five popular feature selection techniques
is compared and analyzed when original dataset is perturbed by
adding, removing or simply resampling the original observations.
Next, the feature selection techniques are used as filter prior to
AdaBoost classifier and their influence on classification accuracy
and Mathews correlation coefficient is compared.",
  address="IEEE",
  booktitle="TENCON 2011 - 2011 IEEE Region 10 Conference",
  chapter="110176",
  doi="10.1109/TENCON.2014.7022309",
  howpublished="electronic, physical medium",
  institution="IEEE",
  year="2014",
  month="october",
  pages="1--4",
  publisher="IEEE",
  type="conference paper"
}