A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring

Zhang Kui, Andrew Ball, Gu Fengshou, Li Yuhua

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Feature selection method has become the focus of research in the area of engineering data processing where there exists a large amount of high-dimensional data from the high-frequency acquisition system. For high-dimensional data processing, engineers often resort to feature extraction methods and statistical theories to convert the original features into new features. However, the converted data always lose the engineering meaning of the original features and the choice and use of conversion methods are challenging. In this paper, a hybrid feature selection model is presented to select the most significant input features from all potentially relevant features. The algorithm combines a filter model with a wrapper model. In the filter model, four variable ranking methods are used to pre-rank the candidate features. These four methods including Pearson correlation coefficient, Relief algorithm, Fisher score and Class separability, measure features from various angles, which leads to different ranking results. Therefore, a weighted voting scheme is introduced to re-rank features based on the degree of significance of the four methods on the classification error rate of Radial Basis Function (RBF) classifier. In wrapper model, a Binary Search (BS) method and a Sequential Backward Search (SBS) method are utilized to minimize the number of relevant features when promising to keep the classification error rate of RBF classifier below a given threshold. To demonstrate the potential of applying the method to large-scale engineering data processing, a case study is conducted.

LanguageEnglish
Title of host publication 2007 IEEE International Conference on Automation Science and Engineering
PublisherIEEE
Pages424-429
Number of pages6
ISBN (Electronic)9781424411542
ISBN (Print)9781424411535
DOIs
Publication statusPublished - 8 Oct 2007
Externally publishedYes
Event3rd IEEE International Conference on Automation Science and Engineering - Scottsdale, United States
Duration: 22 Sep 200725 Sep 2007
Conference number: 3

Conference

Conference3rd IEEE International Conference on Automation Science and Engineering
Abbreviated titleCASE 2007
CountryUnited States
CityScottsdale
Period22/09/0725/09/07

Fingerprint

Condition monitoring
Machinery
Feature extraction
Classifiers
Engineers

Cite this

Kui, Z., Ball, A., Fengshou, G., & Yuhua, L. (2007). A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. In 2007 IEEE International Conference on Automation Science and Engineering (pp. 424-429). [4341697] IEEE. https://doi.org/10.1109/COASE.2007.4341697
Kui, Zhang ; Ball, Andrew ; Fengshou, Gu ; Yuhua, Li. / A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. 2007 IEEE International Conference on Automation Science and Engineering. IEEE, 2007. pp. 424-429
@inproceedings{6859339e9f4f4301ba015e1013522270,
title = "A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring",
abstract = "Feature selection method has become the focus of research in the area of engineering data processing where there exists a large amount of high-dimensional data from the high-frequency acquisition system. For high-dimensional data processing, engineers often resort to feature extraction methods and statistical theories to convert the original features into new features. However, the converted data always lose the engineering meaning of the original features and the choice and use of conversion methods are challenging. In this paper, a hybrid feature selection model is presented to select the most significant input features from all potentially relevant features. The algorithm combines a filter model with a wrapper model. In the filter model, four variable ranking methods are used to pre-rank the candidate features. These four methods including Pearson correlation coefficient, Relief algorithm, Fisher score and Class separability, measure features from various angles, which leads to different ranking results. Therefore, a weighted voting scheme is introduced to re-rank features based on the degree of significance of the four methods on the classification error rate of Radial Basis Function (RBF) classifier. In wrapper model, a Binary Search (BS) method and a Sequential Backward Search (SBS) method are utilized to minimize the number of relevant features when promising to keep the classification error rate of RBF classifier below a given threshold. To demonstrate the potential of applying the method to large-scale engineering data processing, a case study is conducted.",
keywords = "voting, machinery, condition mointoring, data engineering, data processing, error statistics, feature extraction, statistical analysis, weighted voting system, hybrid feature selection model, pearson correlation coefficient, relief algorithm",
author = "Zhang Kui and Andrew Ball and Gu Fengshou and Li Yuhua",
year = "2007",
month = "10",
day = "8",
doi = "10.1109/COASE.2007.4341697",
language = "English",
isbn = "9781424411535",
pages = "424--429",
booktitle = "2007 IEEE International Conference on Automation Science and Engineering",
publisher = "IEEE",

}

Kui, Z, Ball, A, Fengshou, G & Yuhua, L 2007, A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. in 2007 IEEE International Conference on Automation Science and Engineering., 4341697, IEEE, pp. 424-429, 3rd IEEE International Conference on Automation Science and Engineering, Scottsdale, United States, 22/09/07. https://doi.org/10.1109/COASE.2007.4341697

A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. / Kui, Zhang; Ball, Andrew; Fengshou, Gu; Yuhua, Li.

2007 IEEE International Conference on Automation Science and Engineering. IEEE, 2007. p. 424-429 4341697.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring

AU - Kui, Zhang

AU - Ball, Andrew

AU - Fengshou, Gu

AU - Yuhua, Li

PY - 2007/10/8

Y1 - 2007/10/8

N2 - Feature selection method has become the focus of research in the area of engineering data processing where there exists a large amount of high-dimensional data from the high-frequency acquisition system. For high-dimensional data processing, engineers often resort to feature extraction methods and statistical theories to convert the original features into new features. However, the converted data always lose the engineering meaning of the original features and the choice and use of conversion methods are challenging. In this paper, a hybrid feature selection model is presented to select the most significant input features from all potentially relevant features. The algorithm combines a filter model with a wrapper model. In the filter model, four variable ranking methods are used to pre-rank the candidate features. These four methods including Pearson correlation coefficient, Relief algorithm, Fisher score and Class separability, measure features from various angles, which leads to different ranking results. Therefore, a weighted voting scheme is introduced to re-rank features based on the degree of significance of the four methods on the classification error rate of Radial Basis Function (RBF) classifier. In wrapper model, a Binary Search (BS) method and a Sequential Backward Search (SBS) method are utilized to minimize the number of relevant features when promising to keep the classification error rate of RBF classifier below a given threshold. To demonstrate the potential of applying the method to large-scale engineering data processing, a case study is conducted.

AB - Feature selection method has become the focus of research in the area of engineering data processing where there exists a large amount of high-dimensional data from the high-frequency acquisition system. For high-dimensional data processing, engineers often resort to feature extraction methods and statistical theories to convert the original features into new features. However, the converted data always lose the engineering meaning of the original features and the choice and use of conversion methods are challenging. In this paper, a hybrid feature selection model is presented to select the most significant input features from all potentially relevant features. The algorithm combines a filter model with a wrapper model. In the filter model, four variable ranking methods are used to pre-rank the candidate features. These four methods including Pearson correlation coefficient, Relief algorithm, Fisher score and Class separability, measure features from various angles, which leads to different ranking results. Therefore, a weighted voting scheme is introduced to re-rank features based on the degree of significance of the four methods on the classification error rate of Radial Basis Function (RBF) classifier. In wrapper model, a Binary Search (BS) method and a Sequential Backward Search (SBS) method are utilized to minimize the number of relevant features when promising to keep the classification error rate of RBF classifier below a given threshold. To demonstrate the potential of applying the method to large-scale engineering data processing, a case study is conducted.

KW - voting

KW - machinery

KW - condition mointoring

KW - data engineering

KW - data processing

KW - error statistics

KW - feature extraction

KW - statistical analysis

KW - weighted voting system

KW - hybrid feature selection model

KW - pearson correlation coefficient

KW - relief algorithm

UR - http://www.scopus.com/inward/record.url?scp=44449142575&partnerID=8YFLogxK

U2 - 10.1109/COASE.2007.4341697

DO - 10.1109/COASE.2007.4341697

M3 - Conference contribution

SN - 9781424411535

SP - 424

EP - 429

BT - 2007 IEEE International Conference on Automation Science and Engineering

PB - IEEE

ER -

Kui Z, Ball A, Fengshou G, Yuhua L. A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. In 2007 IEEE International Conference on Automation Science and Engineering. IEEE. 2007. p. 424-429. 4341697 https://doi.org/10.1109/COASE.2007.4341697