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

Kui Zhang, Andrew Ball, Fengshou Gu, Yuhua Li

    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 publicationUnknown Host Publication
    Pages174-179
    Number of pages6
    Publication statusPublished - 2007
    Event2007 IEEE INTERNATIONAL CONFERENCE ON AUTOMATION SCIENCE AND ENGINEERING, VOLS 1-3 -
    Duration: 1 Jan 2007 → …

    Conference

    Conference2007 IEEE INTERNATIONAL CONFERENCE ON AUTOMATION SCIENCE AND ENGINEERING, VOLS 1-3
    Period1/01/07 → …

    Fingerprint

    Condition monitoring
    Machinery
    Feature extraction
    Classifiers
    Engineers

    Cite this

    Zhang, K., Ball, A., Gu, F., & Li, Y. (2007). A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. In Unknown Host Publication (pp. 174-179)
    Zhang, Kui ; Ball, Andrew ; Gu, Fengshou ; Li, Yuhua. / A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. Unknown Host Publication. 2007. pp. 174-179
    @inproceedings{20062984d1404797afe67240026bc4dc,
    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.",
    author = "Kui Zhang and Andrew Ball and Fengshou Gu and Yuhua Li",
    note = "IEEE International Conference on Automation Science and Engineering, Scottsdale, AZ, SEP 22-25, 2007",
    year = "2007",
    language = "English",
    pages = "174--179",
    booktitle = "Unknown Host Publication",

    }

    Zhang, K, Ball, A, Gu, F & Li, Y 2007, A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. in Unknown Host Publication. pp. 174-179, 2007 IEEE INTERNATIONAL CONFERENCE ON AUTOMATION SCIENCE AND ENGINEERING, VOLS 1-3, 1/01/07.

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

    Unknown Host Publication. 2007. p. 174-179.

    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 - Zhang, Kui

    AU - Ball, Andrew

    AU - Gu, Fengshou

    AU - Li, Yuhua

    N1 - IEEE International Conference on Automation Science and Engineering, Scottsdale, AZ, SEP 22-25, 2007

    PY - 2007

    Y1 - 2007

    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.

    M3 - Conference contribution

    SP - 174

    EP - 179

    BT - Unknown Host Publication

    ER -

    Zhang K, Ball A, Gu F, Li Y. A hybrid model with a weighted voting scheme for feature selection in machinery condition monitoring. In Unknown Host Publication. 2007. p. 174-179