Building a memetic algorithm based support vector machine for imbalaced classification

Wu Mingnan, Junzo Watada, Zuwarie Ibrahim, Marzuki Khalid

    研究成果: Conference contribution

    抄録

    Classification analysis is one of core research topics in pattern recognition field. According to the distribution of samples, algorithms like artificial network (ANN) and support vector machine (SVM) have been proposed to perform binary classification. But these traditional classification algorithms hardly work well for imbalanced dataset. In this study, a novel model on the basis of memetic algorithm (MA) and support vector machine (SVM) is proposed to perform the classification for large imbalanced dataset. It is named MSVC (memetic support vector classification) model. Memetic Algorithm is recently proposed and used as a heuristic framework for the large imbalanced classification. Because of the high performance of SVM in balanced binary classification, support vector classification (SVC) is combined with MA to improve the classification accuracy. G-mean is used to check the final result. Compared with some conventional models, the results showed that this model is a proper alternative for imbalanced dataset classification, and it expends the applications of memetic algorithm.

    元の言語English
    ホスト出版物のタイトルProceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011
    ページ389-392
    ページ数4
    DOI
    出版物ステータスPublished - 2011
    イベント5th International Conference on Genetic and Evolutionary Computing, ICGEC2011 - Xiamen
    継続期間: 2011 8 292011 9 1

    Other

    Other5th International Conference on Genetic and Evolutionary Computing, ICGEC2011
    Xiamen
    期間11/8/2911/9/1

    Fingerprint

    Support vector machines
    Pattern recognition

    ASJC Scopus subject areas

    • Computational Theory and Mathematics
    • Computer Science Applications

    これを引用

    Mingnan, W., Watada, J., Ibrahim, Z., & Khalid, M. (2011). Building a memetic algorithm based support vector machine for imbalaced classification. : Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011 (pp. 389-392). [6042808] https://doi.org/10.1109/ICGEC.2011.95

    Building a memetic algorithm based support vector machine for imbalaced classification. / Mingnan, Wu; Watada, Junzo; Ibrahim, Zuwarie; Khalid, Marzuki.

    Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011. 2011. p. 389-392 6042808.

    研究成果: Conference contribution

    Mingnan, W, Watada, J, Ibrahim, Z & Khalid, M 2011, Building a memetic algorithm based support vector machine for imbalaced classification. : Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011., 6042808, pp. 389-392, 5th International Conference on Genetic and Evolutionary Computing, ICGEC2011, Xiamen, 11/8/29. https://doi.org/10.1109/ICGEC.2011.95
    Mingnan W, Watada J, Ibrahim Z, Khalid M. Building a memetic algorithm based support vector machine for imbalaced classification. : Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011. 2011. p. 389-392. 6042808 https://doi.org/10.1109/ICGEC.2011.95
    Mingnan, Wu ; Watada, Junzo ; Ibrahim, Zuwarie ; Khalid, Marzuki. / Building a memetic algorithm based support vector machine for imbalaced classification. Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011. 2011. pp. 389-392
    @inproceedings{93b871376ffa4dc6b0ce05d4319afdd4,
    title = "Building a memetic algorithm based support vector machine for imbalaced classification",
    abstract = "Classification analysis is one of core research topics in pattern recognition field. According to the distribution of samples, algorithms like artificial network (ANN) and support vector machine (SVM) have been proposed to perform binary classification. But these traditional classification algorithms hardly work well for imbalanced dataset. In this study, a novel model on the basis of memetic algorithm (MA) and support vector machine (SVM) is proposed to perform the classification for large imbalanced dataset. It is named MSVC (memetic support vector classification) model. Memetic Algorithm is recently proposed and used as a heuristic framework for the large imbalanced classification. Because of the high performance of SVM in balanced binary classification, support vector classification (SVC) is combined with MA to improve the classification accuracy. G-mean is used to check the final result. Compared with some conventional models, the results showed that this model is a proper alternative for imbalanced dataset classification, and it expends the applications of memetic algorithm.",
    keywords = "Classification on imbalanced dataset, Memetic algorithm (MA), Memetic support vector classification (MSVC), Support vector machine (SVM)",
    author = "Wu Mingnan and Junzo Watada and Zuwarie Ibrahim and Marzuki Khalid",
    year = "2011",
    doi = "10.1109/ICGEC.2011.95",
    language = "English",
    isbn = "9780769544496",
    pages = "389--392",
    booktitle = "Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011",

    }

    TY - GEN

    T1 - Building a memetic algorithm based support vector machine for imbalaced classification

    AU - Mingnan, Wu

    AU - Watada, Junzo

    AU - Ibrahim, Zuwarie

    AU - Khalid, Marzuki

    PY - 2011

    Y1 - 2011

    N2 - Classification analysis is one of core research topics in pattern recognition field. According to the distribution of samples, algorithms like artificial network (ANN) and support vector machine (SVM) have been proposed to perform binary classification. But these traditional classification algorithms hardly work well for imbalanced dataset. In this study, a novel model on the basis of memetic algorithm (MA) and support vector machine (SVM) is proposed to perform the classification for large imbalanced dataset. It is named MSVC (memetic support vector classification) model. Memetic Algorithm is recently proposed and used as a heuristic framework for the large imbalanced classification. Because of the high performance of SVM in balanced binary classification, support vector classification (SVC) is combined with MA to improve the classification accuracy. G-mean is used to check the final result. Compared with some conventional models, the results showed that this model is a proper alternative for imbalanced dataset classification, and it expends the applications of memetic algorithm.

    AB - Classification analysis is one of core research topics in pattern recognition field. According to the distribution of samples, algorithms like artificial network (ANN) and support vector machine (SVM) have been proposed to perform binary classification. But these traditional classification algorithms hardly work well for imbalanced dataset. In this study, a novel model on the basis of memetic algorithm (MA) and support vector machine (SVM) is proposed to perform the classification for large imbalanced dataset. It is named MSVC (memetic support vector classification) model. Memetic Algorithm is recently proposed and used as a heuristic framework for the large imbalanced classification. Because of the high performance of SVM in balanced binary classification, support vector classification (SVC) is combined with MA to improve the classification accuracy. G-mean is used to check the final result. Compared with some conventional models, the results showed that this model is a proper alternative for imbalanced dataset classification, and it expends the applications of memetic algorithm.

    KW - Classification on imbalanced dataset

    KW - Memetic algorithm (MA)

    KW - Memetic support vector classification (MSVC)

    KW - Support vector machine (SVM)

    UR - http://www.scopus.com/inward/record.url?scp=80455137177&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=80455137177&partnerID=8YFLogxK

    U2 - 10.1109/ICGEC.2011.95

    DO - 10.1109/ICGEC.2011.95

    M3 - Conference contribution

    AN - SCOPUS:80455137177

    SN - 9780769544496

    SP - 389

    EP - 392

    BT - Proceedings - 2011 5th International Conference on Genetic and Evolutionary Computing, ICGEC 2011

    ER -