A two-step supervised learning artificial neural network for imbalanced dataset problems

Asrul Adam, Zuwairie Ibrahim, Mohd Ibrahim Shapiai, Lim Chun Chew, Lee Wen Jau, Marzuki Khalid, Junzo Watada

    Research output: Contribution to journalArticle

    6 Citations (Scopus)

    Abstract

    In this paper, a two-step supervised learning algorithm of a single layer feedforward Artificial Neural Network (ANN) is proposed for solving imbalanced dataset problems. Levenberg Marquart backpropagation learning algorithm is utilized in the first step learning, while the second step learning mechanism is introduced by optimizing the decision threshold of the step function at the output layer of ANN using particle swarm optimization (PSO). After all the steps learning are accomplished, the best weights and decision threshold value are obtained to be used for testing process. Several imbalanced datasets, which are available in UCI Machine Learning Repository, are chosen as case study. The prediction performance is assessed by Geometric Mean (G-mean), which is a standard measure to indicate the efficiency of classifier for imbalanced datasets. Based on the experimental results, the proposed method is able to provide good G-mean value compared with the conventional ANN approaches.

    Original languageEnglish
    Pages (from-to)3163-3172
    Number of pages10
    JournalInternational Journal of Innovative Computing, Information and Control
    Volume8
    Issue number5 A
    Publication statusPublished - 2012 May

    Fingerprint

    Supervised learning
    Supervised Learning
    Artificial Neural Network
    Geometric mean
    Neural networks
    Learning algorithms
    Learning Algorithm
    Network Optimization
    Step function
    Backpropagation algorithms
    Back-propagation Algorithm
    Performance Prediction
    Feedforward
    Threshold Value
    Mean Value
    Repository
    Particle swarm optimization (PSO)
    Particle Swarm Optimization
    Learning systems
    Machine Learning

    Keywords

    • Articial neural network
    • Decision threshold
    • Imbalanced dataset problem
    • Machine learning
    • Particle swarm opti-mization
    • Single layer feedforward neural network
    • Two-class classication

    ASJC Scopus subject areas

    • Computational Theory and Mathematics
    • Information Systems
    • Software
    • Theoretical Computer Science

    Cite this

    Adam, A., Ibrahim, Z., Shapiai, M. I., Chew, L. C., Jau, L. W., Khalid, M., & Watada, J. (2012). A two-step supervised learning artificial neural network for imbalanced dataset problems. International Journal of Innovative Computing, Information and Control, 8(5 A), 3163-3172.

    A two-step supervised learning artificial neural network for imbalanced dataset problems. / Adam, Asrul; Ibrahim, Zuwairie; Shapiai, Mohd Ibrahim; Chew, Lim Chun; Jau, Lee Wen; Khalid, Marzuki; Watada, Junzo.

    In: International Journal of Innovative Computing, Information and Control, Vol. 8, No. 5 A, 05.2012, p. 3163-3172.

    Research output: Contribution to journalArticle

    Adam, A, Ibrahim, Z, Shapiai, MI, Chew, LC, Jau, LW, Khalid, M & Watada, J 2012, 'A two-step supervised learning artificial neural network for imbalanced dataset problems', International Journal of Innovative Computing, Information and Control, vol. 8, no. 5 A, pp. 3163-3172.
    Adam, Asrul ; Ibrahim, Zuwairie ; Shapiai, Mohd Ibrahim ; Chew, Lim Chun ; Jau, Lee Wen ; Khalid, Marzuki ; Watada, Junzo. / A two-step supervised learning artificial neural network for imbalanced dataset problems. In: International Journal of Innovative Computing, Information and Control. 2012 ; Vol. 8, No. 5 A. pp. 3163-3172.
    @article{d42b8a5887b04c94b99129a0ac0c901f,
    title = "A two-step supervised learning artificial neural network for imbalanced dataset problems",
    abstract = "In this paper, a two-step supervised learning algorithm of a single layer feedforward Artificial Neural Network (ANN) is proposed for solving imbalanced dataset problems. Levenberg Marquart backpropagation learning algorithm is utilized in the first step learning, while the second step learning mechanism is introduced by optimizing the decision threshold of the step function at the output layer of ANN using particle swarm optimization (PSO). After all the steps learning are accomplished, the best weights and decision threshold value are obtained to be used for testing process. Several imbalanced datasets, which are available in UCI Machine Learning Repository, are chosen as case study. The prediction performance is assessed by Geometric Mean (G-mean), which is a standard measure to indicate the efficiency of classifier for imbalanced datasets. Based on the experimental results, the proposed method is able to provide good G-mean value compared with the conventional ANN approaches.",
    keywords = "Articial neural network, Decision threshold, Imbalanced dataset problem, Machine learning, Particle swarm opti-mization, Single layer feedforward neural network, Two-class classication",
    author = "Asrul Adam and Zuwairie Ibrahim and Shapiai, {Mohd Ibrahim} and Chew, {Lim Chun} and Jau, {Lee Wen} and Marzuki Khalid and Junzo Watada",
    year = "2012",
    month = "5",
    language = "English",
    volume = "8",
    pages = "3163--3172",
    journal = "International Journal of Innovative Computing, Information and Control",
    issn = "1349-4198",
    publisher = "IJICIC Editorial Office",
    number = "5 A",

    }

    TY - JOUR

    T1 - A two-step supervised learning artificial neural network for imbalanced dataset problems

    AU - Adam, Asrul

    AU - Ibrahim, Zuwairie

    AU - Shapiai, Mohd Ibrahim

    AU - Chew, Lim Chun

    AU - Jau, Lee Wen

    AU - Khalid, Marzuki

    AU - Watada, Junzo

    PY - 2012/5

    Y1 - 2012/5

    N2 - In this paper, a two-step supervised learning algorithm of a single layer feedforward Artificial Neural Network (ANN) is proposed for solving imbalanced dataset problems. Levenberg Marquart backpropagation learning algorithm is utilized in the first step learning, while the second step learning mechanism is introduced by optimizing the decision threshold of the step function at the output layer of ANN using particle swarm optimization (PSO). After all the steps learning are accomplished, the best weights and decision threshold value are obtained to be used for testing process. Several imbalanced datasets, which are available in UCI Machine Learning Repository, are chosen as case study. The prediction performance is assessed by Geometric Mean (G-mean), which is a standard measure to indicate the efficiency of classifier for imbalanced datasets. Based on the experimental results, the proposed method is able to provide good G-mean value compared with the conventional ANN approaches.

    AB - In this paper, a two-step supervised learning algorithm of a single layer feedforward Artificial Neural Network (ANN) is proposed for solving imbalanced dataset problems. Levenberg Marquart backpropagation learning algorithm is utilized in the first step learning, while the second step learning mechanism is introduced by optimizing the decision threshold of the step function at the output layer of ANN using particle swarm optimization (PSO). After all the steps learning are accomplished, the best weights and decision threshold value are obtained to be used for testing process. Several imbalanced datasets, which are available in UCI Machine Learning Repository, are chosen as case study. The prediction performance is assessed by Geometric Mean (G-mean), which is a standard measure to indicate the efficiency of classifier for imbalanced datasets. Based on the experimental results, the proposed method is able to provide good G-mean value compared with the conventional ANN approaches.

    KW - Articial neural network

    KW - Decision threshold

    KW - Imbalanced dataset problem

    KW - Machine learning

    KW - Particle swarm opti-mization

    KW - Single layer feedforward neural network

    KW - Two-class classication

    UR - http://www.scopus.com/inward/record.url?scp=84860825860&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84860825860&partnerID=8YFLogxK

    M3 - Article

    VL - 8

    SP - 3163

    EP - 3172

    JO - International Journal of Innovative Computing, Information and Control

    JF - International Journal of Innovative Computing, Information and Control

    SN - 1349-4198

    IS - 5 A

    ER -