Improved weighted least squares support vector machines algorithm and its applications in spectroscopic quantitative analysis

被引：0
|
作者：
Lu Jian-Feng [1 ]
Dai Lian-Kui [1 ]
机构：
[1] Zhejiang Univ, Natl Key Lab Ind Control Technol, Hangzhou 310027, Peoples R China
来源：
CHINESE JOURNAL OF ANALYTICAL CHEMISTRY | 2007年 / 35卷 / 03期
关键词：
support vector machines; anomaly detection; robust modeling; near-infrared spectroscopy;
D O I：
暂无
中图分类号：
O65 [分析化学];
学科分类号：
070302 ; 081704 ;
摘要：
An improved algorithm of weighted least squares support vector machines (WLS-SVM) is proposed to overcome the negatine influence of abnornmal traningsamples on calibration models. The improved algorithm solves the iterative convergence problem in the original algorithm and has been used in spectroscopic quantitative analysis. Compared to the original algorithm, the experimental results show that the new algorithm remarkably improves model robustness and the ability to detect abnormal samples.
引用
收藏
页码：340 / 344
页数：5
相关论文
共 10 条
  • [1] Application of LS-SVM to non-linear phenomena in NIR spectroscopy: development of a robust and portable sensor for acidity prediction in grapes
    Chauchard, F
    Cogdill, R
    Roussel, S
    Roger, JM
    Bellon-Maurel, V
    [J]. CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2004, 71 (02) : 141 - 150
  • [2] Chu XL, 2005, SPECTROSC SPECT ANAL, V25, P886
  • [3] Du SX, 2004, CHINESE J ANAL CHEM, V32, P1227
  • [4] LIANG YZ, 1996, CHEMISTRY, P26
  • [5] Liu SH, 2005, SPECTROSC SPECT ANAL, V25, P878
  • [6] Weighted least squares support vector machines: robustness and sparse approximation
    Suykens, JAK
    De Brabanter, J
    Lukas, L
    Vandewalle, J
    [J]. NEUROCOMPUTING, 2002, 48 : 85 - 105
  • [7] Training multilayer perceptron classifiers based on a modified support vector method
    Suykens, JAK
    Vandewalle, J
    [J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1999, 10 (04): : 907 - 911
  • [8] Comparing support vector machines to PLS for spectral regression applications
    Thissen, U
    Pepers, M
    Üstün, B
    Melssen, WJ
    Buydens, LMC
    [J]. CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2004, 73 (02) : 169 - 179
  • [9] Vapnik V., 1998, STAT LEARNING THEORY, V1, P2
  • [10] Discriminating the genuineness of Chinese medicines with least squares support vector machines
    Yu, K
    Cheng, YY
    [J]. CHINESE JOURNAL OF ANALYTICAL CHEMISTRY, 2006, 34 (04) : 561 - 564
1