Variable Selection Method of NIR Spectroscopy Based on Least Angle Regression and GA-PLS
YAN Sheng-ke1, YANG Hui-hua1, 2*, HU Bai-chao1, REN Chao-chao1, LIU Zhen-bing1
1. College of Electronic Engineering and Automation, Guilin University of Electronic Technology, Guilin 541004, China
2. College of Automation, Beijing University of Posts & Telecommunications, Beijing 100876, China
Abstract：Near infrared (NIR) spectra usually have many wavelength variables. Direct or indirect variable selection is crucial to improve the stability and prediction performance of a model. Least angle regression (LAR) is a relatively new and efficient machine learning algorithm for regression analysis and variable selection. By combining LAR and genetic algorithm-partial least square (GA-PLS) algorithm, a wavelength selection method is proposed in this paper for spectral modeling applications, which can effectively screen a few wavelength points. Firstly, LAR is used to eliminate the multiple-collinearity among variables in the full spectrum region and get a reduced set of features, then GA-PLS is introduced to select the variables from the reduced set of features to achieve the purpose of further dimension reduction. In order to verify the validity of it, the method is carried out by making regression analysis on the NIR spectroscopy of tablets and gasoline. The pre-processing results of original spectra are used to select the variables and have modeled on the content of active ingredients (Tablets) and C10 (Gasoline). As a result, the optimal number of variables are just 7 in both of applications, and the predictive decision coefficient (R2p) reached 0.933 9 and 0.951 9 respectively. Moreover, by comparing with the full spectrum, elimination of uninformative variables (UVE) and successive projection algorithm (SPA) model, the result shows that this method needs less wavelength points and have more excellent in R2p and root mean square error of predication (RMSEP). Therefore, LAR combined with GA-PLS not only can picks out information variables from NIR spectroscopy to reduce the variable number for modeling and improve the prediction accuracy, but also has a better interpretation of the model. The method can provide as effective wavelength selection tool for designing of special spectrometer in particular area.
Key words：Near-infrared spectroscopy; Least angle regression; Variable selection; Collinearity; Genetic algorithm-partial least square
颜胜科，杨辉华，胡百超，任超超，刘振丙. 基于最小角回归与GA-PLS的NIR光谱变量选择方法[J]. 光谱学与光谱分析, 2017, 37(06): 1733-1738.
YAN Sheng-ke, YANG Hui-hua, HU Bai-chao, REN Chao-chao, LIU Zhen-bing. Variable Selection Method of NIR Spectroscopy Based on Least Angle Regression and GA-PLS. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2017, 37(06): 1733-1738.
[1] CHU Xiao-li, LU Wan-zhen(褚小立, 陆婉珍). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2014, 34(10): 2595.
[2] Centner V, Massart D L, Noord O E, et al. Analytical Chemistry, 1996, 68(21): 3851.
[3] Araújo M C U, Saldanha T C B, Galvo R K H, et al. Chemometrics & Intelligent Laboratory Systems, 2001, 57(2): 65.
[4] Nrgaard L, Saudland A, Wagner J, et al. Applied Spectroscopy, 2000, 54(3): 413.
[5] Efron B, Hastie T, Johnstone I, et al. The Annals of Statistics, 2004, 32(2): 407.
[6] Liu Cong, Simon X Yang, Deng Lie. Expert Systems with Applications, 2015, 42(22): 8497.
[7] Zhang Long, Li Kang. Automatica, 2015, 53: 94.
[8] Leardi R. Journal of Chemometrics, 2000, 14(5-6): 643.
[9] Liu Huan, Motoda H, Setiono R, et al. Hyderabad, 2010, 10: 4.
[10] Norris K H, Ritchie G E. Journal of Pharmaceutical & Biomedical Analysis, 2008, 48(3): 1037.
[11] CHU Xiao-li(褚小立). Molecular Spectroscopy Analytical Technology Combined with Chemometrics and Its Applications(化学计量学方法与分子光谱分析技术). Beijing: Chemical Industry Press(北京：化学工业出版社), 2011.
[12] LU Wan-zhen(陆婉珍). Modern Near Infrared Spectroscopy Analytical Technology(现代近红外光谱分析技术). Beijing: China Petrochemical Pres(北京：中国石化出版社), 2007.
[13] SONG Hao(宋 浩). China Petroleum and Chemical Standard and Quality(中国石油和化工标准与质量), 2013, (9): 15.
