• DocumentCode
    475976
  • Title

    A self-adapting regression SVM model and its application on middle managers performance evaluation

  • Author

    Yang, Shao-mei

  • Author_Institution
    Econ. & Manage. Dept., North China Electr. Power Univ., Beijing
  • Volume
    2
  • fYear
    2008
  • fDate
    12-15 July 2008
  • Firstpage
    645
  • Lastpage
    648
  • Abstract
    It is very important that middle managers work performance affect the survival and development of enterprises, carry on reasonable, scientific and effective performance evaluation, and encourage them to give full play to the initiative and creativity, which is the reliable protection of achieving the enterprises short-term and long-term objectives. Based on the SVM algorithm theory, this paper establish the middle managers performance evaluation model based on self-adapting regression SVM, through establishing middle managers performance index system, use several SVM classifier series portfolio, solve the problems that the training samplespsila category and quantity are imbalance, and the data is interfered, realize the performance evaluation to middle managers. Experimental results show that the method improved the middle managers performance evaluation accuracy and efficiency.
  • Keywords
    appraisal; personnel; regression analysis; support vector machines; SVM classifier series portfolio; middle managers performance evaluation; performance evaluation model; self-adapting regression SVM model; Conference management; Cybernetics; Electronic mail; Energy management; Kernel; Machine learning; Management training; Power generation economics; Support vector machine classification; Support vector machines; Self-adapting; middle managers; performance evaluation; regression SVM;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Machine Learning and Cybernetics, 2008 International Conference on
  • Conference_Location
    Kunming
  • Print_ISBN
    978-1-4244-2095-7
  • Electronic_ISBN
    978-1-4244-2096-4
  • Type

    conf

  • DOI
    10.1109/ICMLC.2008.4620484
  • Filename
    4620484