Feature selection in growing dimensional space for classification based on neural incremental attribute learning

Ting Wang; Sheng Uei Guan; Fei Liu

doi:10.1007/978-3-642-40618-8_66

Feature selection in growing dimensional space for classification based on neural incremental attribute learning

Ting Wang^*, Sheng Uei Guan, Fei Liu

^*Corresponding author for this work

Department of Computing

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

1 Citation (Scopus)

Abstract

As a novel machine learning strategy, Incremental Attribute Learning (IAL) gradually trains features in increments. Algorithms using IAL often exhibit better performance than some other approaches in previous studies. However, in previous research, feature selection for IAL usually employed wrapper approaches, which are low in efficiency. Moreover, IAL has its own characteristics like growing dimensional feature space, where traditional stable feature selection methods cannot perform well. Therefore, new feature selection methods must be designed for IAL. In this paper, a dynamic feature selection approach based on feature ordering is presented. In the meanwhile, metrics and algorithms for getting optimal feature ordering and selection are also interpreted in this paper.

Original language	English
Title of host publication	Proceedings of the 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013
Publisher	Springer Verlag
Pages	501-507
Number of pages	7
Edition	VOL. 1
ISBN (Print)	9783642406171
DOIs	https://doi.org/10.1007/978-3-642-40618-8_66
Publication status	Published - 2014
Event	9th International Symposium on Linear Drives for Industry Applications, LDIA 2013 - Hangzhou, China Duration: 7 Jul 2013 → 10 Jul 2013

Publication series

Name	Lecture Notes in Electrical Engineering
Number	VOL. 1
Volume	270 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	9th International Symposium on Linear Drives for Industry Applications, LDIA 2013
Country/Territory	China
City	Hangzhou
Period	7/07/13 → 10/07/13

Keywords

Feature ordering
Feature selection
Incremental attribute learning
Neural networks
Pattern classification

Access to Document

10.1007/978-3-642-40618-8_66

Cite this

Wang, T., Guan, S. U., & Liu, F. (2014). Feature selection in growing dimensional space for classification based on neural incremental attribute learning. In Proceedings of the 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013 (VOL. 1 ed., pp. 501-507). (Lecture Notes in Electrical Engineering; Vol. 270 LNEE, No. VOL. 1). Springer Verlag. https://doi.org/10.1007/978-3-642-40618-8_66

@inproceedings{904d0d53cb6745dd9c9119b379387d86,

title = "Feature selection in growing dimensional space for classification based on neural incremental attribute learning",

abstract = "As a novel machine learning strategy, Incremental Attribute Learning (IAL) gradually trains features in increments. Algorithms using IAL often exhibit better performance than some other approaches in previous studies. However, in previous research, feature selection for IAL usually employed wrapper approaches, which are low in efficiency. Moreover, IAL has its own characteristics like growing dimensional feature space, where traditional stable feature selection methods cannot perform well. Therefore, new feature selection methods must be designed for IAL. In this paper, a dynamic feature selection approach based on feature ordering is presented. In the meanwhile, metrics and algorithms for getting optimal feature ordering and selection are also interpreted in this paper.",

keywords = "Feature ordering, Feature selection, Incremental attribute learning, Neural networks, Pattern classification",

author = "Ting Wang and Guan, {Sheng Uei} and Fei Liu",

year = "2014",

doi = "10.1007/978-3-642-40618-8_66",

language = "English",

isbn = "9783642406171",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Verlag",

number = "VOL. 1",

pages = "501--507",

booktitle = "Proceedings of the 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013",

edition = "VOL. 1",

note = "9th International Symposium on Linear Drives for Industry Applications, LDIA 2013 ; Conference date: 07-07-2013 Through 10-07-2013",

}

Wang, T, Guan, SU & Liu, F 2014, Feature selection in growing dimensional space for classification based on neural incremental attribute learning. in Proceedings of the 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013. VOL. 1 edn, Lecture Notes in Electrical Engineering, no. VOL. 1, vol. 270 LNEE, Springer Verlag, pp. 501-507, 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013, Hangzhou, China, 7/07/13. https://doi.org/10.1007/978-3-642-40618-8_66

Feature selection in growing dimensional space for classification based on neural incremental attribute learning. / Wang, Ting; Guan, Sheng Uei; Liu, Fei.
Proceedings of the 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013. VOL. 1. ed. Springer Verlag, 2014. p. 501-507 (Lecture Notes in Electrical Engineering; Vol. 270 LNEE, No. VOL. 1).

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

TY - GEN

T1 - Feature selection in growing dimensional space for classification based on neural incremental attribute learning

AU - Wang, Ting

AU - Guan, Sheng Uei

AU - Liu, Fei

PY - 2014

Y1 - 2014

N2 - As a novel machine learning strategy, Incremental Attribute Learning (IAL) gradually trains features in increments. Algorithms using IAL often exhibit better performance than some other approaches in previous studies. However, in previous research, feature selection for IAL usually employed wrapper approaches, which are low in efficiency. Moreover, IAL has its own characteristics like growing dimensional feature space, where traditional stable feature selection methods cannot perform well. Therefore, new feature selection methods must be designed for IAL. In this paper, a dynamic feature selection approach based on feature ordering is presented. In the meanwhile, metrics and algorithms for getting optimal feature ordering and selection are also interpreted in this paper.

AB - As a novel machine learning strategy, Incremental Attribute Learning (IAL) gradually trains features in increments. Algorithms using IAL often exhibit better performance than some other approaches in previous studies. However, in previous research, feature selection for IAL usually employed wrapper approaches, which are low in efficiency. Moreover, IAL has its own characteristics like growing dimensional feature space, where traditional stable feature selection methods cannot perform well. Therefore, new feature selection methods must be designed for IAL. In this paper, a dynamic feature selection approach based on feature ordering is presented. In the meanwhile, metrics and algorithms for getting optimal feature ordering and selection are also interpreted in this paper.

KW - Feature ordering

KW - Feature selection

KW - Incremental attribute learning

KW - Neural networks

KW - Pattern classification

UR - http://www.scopus.com/inward/record.url?scp=84958539812&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-40618-8_66

DO - 10.1007/978-3-642-40618-8_66

M3 - Conference Proceeding

AN - SCOPUS:84958539812

SN - 9783642406171

T3 - Lecture Notes in Electrical Engineering

SP - 501

EP - 507

BT - Proceedings of the 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013

PB - Springer Verlag

T2 - 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013

Y2 - 7 July 2013 through 10 July 2013

ER -

Wang T, Guan SU, Liu F. Feature selection in growing dimensional space for classification based on neural incremental attribute learning. In Proceedings of the 9th International Symposium on Linear Drives for Industry Applications, LDIA 2013. VOL. 1 ed. Springer Verlag. 2014. p. 501-507. (Lecture Notes in Electrical Engineering; VOL. 1). doi: 10.1007/978-3-642-40618-8_66

Feature selection in growing dimensional space for classification based on neural incremental attribute learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this