• DocumentCode
    511249
  • Title

    The Individual Credit Evaluation Based on COLS-SVM

  • Author

    Taian, Liu ; Yunjia, Wang ; Wentong, Liu

  • Author_Institution
    Coll. of Environ. & Spatial Inf., China Univ. of Min. & Technol. (CUMT), Xuzhou, China
  • Volume
    1
  • fYear
    2009
  • fDate
    25-27 Dec. 2009
  • Firstpage
    455
  • Lastpage
    457
  • Abstract
    Least squares support vector machine (LS-SVM) has been widely used in engineering practice, using for reference algorithm of combinatorial optimization, this paper puts forward the combinatorial optimization least squares support vector machine algorithm (COLS-SVM). Based on algorithmic analysis of COLS-SVM, it can be used on individual credit evaluation and compared with Lagrange support vector machine (LSVM) and K-nearest neighbor (KNN), the numerical experiment results show that the proposed COLS-SVM algorithm has good classified forecast ability. Individual credit evaluation is a general reflect of individual capital and credit situation, it has great significance and application value.
  • Keywords
    combinatorial mathematics; financial data processing; least squares approximations; mathematics computing; optimisation; support vector machines; COLS-SVM; K-nearest neighbor; Lagrange support vector machine; algorithmic analysis; combinatorial optimization least squares support vector machine algorithm; individual credit evaluation; Algorithm design and analysis; Application software; Computer applications; Educational institutions; Equations; Lagrangian functions; Least squares methods; Optimization methods; Support vector machine classification; Support vector machines; COLS-SVM; Individual credit characteristic data; Individual credit evaluation; LS-SVM;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer Science-Technology and Applications, 2009. IFCSTA '09. International Forum on
  • Conference_Location
    Chongqing
  • Print_ISBN
    978-0-7695-3930-0
  • Electronic_ISBN
    978-1-4244-5423-5
  • Type

    conf

  • DOI
    10.1109/IFCSTA.2009.117
  • Filename
    5385034