An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization

Yiyuan Chen, Yufeng Wang, Liang Cao, Qun Jin

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

    Abstract

    Feature selection (FS) is one of fundamental data processing techniques in various machine learning algorithms, especially for classification of healthcare data. However, it is a challenging issue due to the large search space. This paper proposed a confidence based and cost effective feature selection method using binary particle swarm optimization, CCFS. First, CCFS improves search effectiveness by developing a new updating mechanism, in which confidence of each feature is explicitly considered, including the correlation between feature and categories, and historically selected frequency of each feature. Second, the classification accuracy, the feature reduction ratio, and the feature cost are comprehensively incorporated into the design of the fitness function. The proposed method has been verified in UCI cancer classification dataset (Lung Cancer). The experimental result shows the effectiveness of the proposed method, in terms of accuracy and feature selection cost.

    Original languageEnglish
    Title of host publicationProceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018
    PublisherInstitute of Electrical and Electronics Engineers Inc.
    Pages703-707
    Number of pages5
    ISBN (Electronic)9781538677438
    DOIs
    Publication statusPublished - 2018 Dec 26
    Event9th International Conference on Information Technology in Medicine and Education, ITME 2018 - Hangzhou, Zhejiang, China
    Duration: 2018 Oct 192018 Oct 21

    Publication series

    NameProceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018

    Conference

    Conference9th International Conference on Information Technology in Medicine and Education, ITME 2018
    CountryChina
    CityHangzhou, Zhejiang
    Period18/10/1918/10/21

    Fingerprint

    Particle swarm optimization (PSO)
    Feature extraction
    Delivery of Health Care
    Costs and Cost Analysis
    cancer
    costs
    confidence
    Costs
    fitness
    Learning algorithms
    Learning systems
    Lung Neoplasms
    learning
    Neoplasms
    Datasets
    Machine Learning

    Keywords

    • Binary Particle Swarm Optimization
    • Feature selection
    • Healthcare data classification
    • Swarm intelligence

    ASJC Scopus subject areas

    • Computer Science Applications
    • Medicine (miscellaneous)
    • Information Systems
    • Health Informatics
    • Education

    Cite this

    Chen, Y., Wang, Y., Cao, L., & Jin, Q. (2018). An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization. In Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018 (pp. 703-707). [8589392] (Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ITME.2018.00160

    An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization. / Chen, Yiyuan; Wang, Yufeng; Cao, Liang; Jin, Qun.

    Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 703-707 8589392 (Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018).

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

    Chen, Y, Wang, Y, Cao, L & Jin, Q 2018, An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization. in Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018., 8589392, Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018, Institute of Electrical and Electronics Engineers Inc., pp. 703-707, 9th International Conference on Information Technology in Medicine and Education, ITME 2018, Hangzhou, Zhejiang, China, 18/10/19. https://doi.org/10.1109/ITME.2018.00160
    Chen Y, Wang Y, Cao L, Jin Q. An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization. In Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 703-707. 8589392. (Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018). https://doi.org/10.1109/ITME.2018.00160
    Chen, Yiyuan ; Wang, Yufeng ; Cao, Liang ; Jin, Qun. / An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization. Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 703-707 (Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018).
    @inproceedings{80d7af3cfbae4cdeb79df955e5613878,
    title = "An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization",
    abstract = "Feature selection (FS) is one of fundamental data processing techniques in various machine learning algorithms, especially for classification of healthcare data. However, it is a challenging issue due to the large search space. This paper proposed a confidence based and cost effective feature selection method using binary particle swarm optimization, CCFS. First, CCFS improves search effectiveness by developing a new updating mechanism, in which confidence of each feature is explicitly considered, including the correlation between feature and categories, and historically selected frequency of each feature. Second, the classification accuracy, the feature reduction ratio, and the feature cost are comprehensively incorporated into the design of the fitness function. The proposed method has been verified in UCI cancer classification dataset (Lung Cancer). The experimental result shows the effectiveness of the proposed method, in terms of accuracy and feature selection cost.",
    keywords = "Binary Particle Swarm Optimization, Feature selection, Healthcare data classification, Swarm intelligence",
    author = "Yiyuan Chen and Yufeng Wang and Liang Cao and Qun Jin",
    year = "2018",
    month = "12",
    day = "26",
    doi = "10.1109/ITME.2018.00160",
    language = "English",
    series = "Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    pages = "703--707",
    booktitle = "Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018",

    }

    TY - GEN

    T1 - An Effective Feature Selection Scheme for Healthcare Data Classification Using Binary Particle Swarm Optimization

    AU - Chen, Yiyuan

    AU - Wang, Yufeng

    AU - Cao, Liang

    AU - Jin, Qun

    PY - 2018/12/26

    Y1 - 2018/12/26

    N2 - Feature selection (FS) is one of fundamental data processing techniques in various machine learning algorithms, especially for classification of healthcare data. However, it is a challenging issue due to the large search space. This paper proposed a confidence based and cost effective feature selection method using binary particle swarm optimization, CCFS. First, CCFS improves search effectiveness by developing a new updating mechanism, in which confidence of each feature is explicitly considered, including the correlation between feature and categories, and historically selected frequency of each feature. Second, the classification accuracy, the feature reduction ratio, and the feature cost are comprehensively incorporated into the design of the fitness function. The proposed method has been verified in UCI cancer classification dataset (Lung Cancer). The experimental result shows the effectiveness of the proposed method, in terms of accuracy and feature selection cost.

    AB - Feature selection (FS) is one of fundamental data processing techniques in various machine learning algorithms, especially for classification of healthcare data. However, it is a challenging issue due to the large search space. This paper proposed a confidence based and cost effective feature selection method using binary particle swarm optimization, CCFS. First, CCFS improves search effectiveness by developing a new updating mechanism, in which confidence of each feature is explicitly considered, including the correlation between feature and categories, and historically selected frequency of each feature. Second, the classification accuracy, the feature reduction ratio, and the feature cost are comprehensively incorporated into the design of the fitness function. The proposed method has been verified in UCI cancer classification dataset (Lung Cancer). The experimental result shows the effectiveness of the proposed method, in terms of accuracy and feature selection cost.

    KW - Binary Particle Swarm Optimization

    KW - Feature selection

    KW - Healthcare data classification

    KW - Swarm intelligence

    UR - http://www.scopus.com/inward/record.url?scp=85061313170&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85061313170&partnerID=8YFLogxK

    U2 - 10.1109/ITME.2018.00160

    DO - 10.1109/ITME.2018.00160

    M3 - Conference contribution

    AN - SCOPUS:85061313170

    T3 - Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018

    SP - 703

    EP - 707

    BT - Proceedings - 9th International Conference on Information Technology in Medicine and Education, ITME 2018

    PB - Institute of Electrical and Electronics Engineers Inc.

    ER -