Classification Using Markov Blanket for Feature Selection

Yifeng Zeng, Jian Luo

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

12 Citations (Scopus)

Abstract

Selecting relevant features is in demand when a large data set is of interest in a classification task. It produces a tractable number of features that are sufficient and possibly improve the classification performance. This paper studies a statistical method of Markov blanket induction algorithm for filtering features and then applies a classifier using the Markov blanket predictors. The Markov blanket contains a minimal subset of relevant features that yields optimal classification performance. We experimentally demonstrate the improved performance of several classifiers using a Markov blanket induction as a feature selection method. In addition, we point out an important assumption behind the Markov blanket induction algorithm and show its effect on the classification performance.
Original languageEnglish
Title of host publicationIEEE International Conference on Granular Computing (GrC '09)
PublisherIEEE
Publication date2009
Pages743-747
ISBN (Print)978-1-4244-4830-2
DOIs
Publication statusPublished - 2009
EventThe 2009 IEEE International Conference of Granular Computing, GrC 2009 - Nanchang, China
Duration: 17 Aug 200919 Aug 2009

Conference

ConferenceThe 2009 IEEE International Conference of Granular Computing, GrC 2009
CountryChina
CityNanchang
Period17/08/200919/08/2009

Fingerprint

Feature extraction
Classifiers
Statistical methods

Cite this

Zeng, Y., & Luo, J. (2009). Classification Using Markov Blanket for Feature Selection. In IEEE International Conference on Granular Computing (GrC '09) (pp. 743-747). IEEE. https://doi.org/10.1109/GRC.2009.5255023
Zeng, Yifeng ; Luo, Jian. / Classification Using Markov Blanket for Feature Selection. IEEE International Conference on Granular Computing (GrC '09). IEEE, 2009. pp. 743-747
@inproceedings{a394f230cdf611de9e10000ea68e967b,
title = "Classification Using Markov Blanket for Feature Selection",
abstract = "Selecting relevant features is in demand when a large data set is of interest in a classification task. It produces a tractable number of features that are sufficient and possibly improve the classification performance. This paper studies a statistical method of Markov blanket induction algorithm for filtering features and then applies a classifier using the Markov blanket predictors. The Markov blanket contains a minimal subset of relevant features that yields optimal classification performance. We experimentally demonstrate the improved performance of several classifiers using a Markov blanket induction as a feature selection method. In addition, we point out an important assumption behind the Markov blanket induction algorithm and show its effect on the classification performance.",
author = "Yifeng Zeng and Jian Luo",
year = "2009",
doi = "10.1109/GRC.2009.5255023",
language = "English",
isbn = "978-1-4244-4830-2",
pages = "743--747",
booktitle = "IEEE International Conference on Granular Computing (GrC '09)",
publisher = "IEEE",
address = "United States",

}

Zeng, Y & Luo, J 2009, Classification Using Markov Blanket for Feature Selection. in IEEE International Conference on Granular Computing (GrC '09). IEEE, pp. 743-747, Nanchang, China, 17/08/2009. https://doi.org/10.1109/GRC.2009.5255023

Classification Using Markov Blanket for Feature Selection. / Zeng, Yifeng; Luo, Jian.

IEEE International Conference on Granular Computing (GrC '09). IEEE, 2009. p. 743-747.

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review

TY - GEN

T1 - Classification Using Markov Blanket for Feature Selection

AU - Zeng, Yifeng

AU - Luo, Jian

PY - 2009

Y1 - 2009

N2 - Selecting relevant features is in demand when a large data set is of interest in a classification task. It produces a tractable number of features that are sufficient and possibly improve the classification performance. This paper studies a statistical method of Markov blanket induction algorithm for filtering features and then applies a classifier using the Markov blanket predictors. The Markov blanket contains a minimal subset of relevant features that yields optimal classification performance. We experimentally demonstrate the improved performance of several classifiers using a Markov blanket induction as a feature selection method. In addition, we point out an important assumption behind the Markov blanket induction algorithm and show its effect on the classification performance.

AB - Selecting relevant features is in demand when a large data set is of interest in a classification task. It produces a tractable number of features that are sufficient and possibly improve the classification performance. This paper studies a statistical method of Markov blanket induction algorithm for filtering features and then applies a classifier using the Markov blanket predictors. The Markov blanket contains a minimal subset of relevant features that yields optimal classification performance. We experimentally demonstrate the improved performance of several classifiers using a Markov blanket induction as a feature selection method. In addition, we point out an important assumption behind the Markov blanket induction algorithm and show its effect on the classification performance.

U2 - 10.1109/GRC.2009.5255023

DO - 10.1109/GRC.2009.5255023

M3 - Article in proceeding

SN - 978-1-4244-4830-2

SP - 743

EP - 747

BT - IEEE International Conference on Granular Computing (GrC '09)

PB - IEEE

ER -

Zeng Y, Luo J. Classification Using Markov Blanket for Feature Selection. In IEEE International Conference on Granular Computing (GrC '09). IEEE. 2009. p. 743-747 https://doi.org/10.1109/GRC.2009.5255023