Genetic Algorithm-Based Wavelength Selection for the Near-Infrared Determination of Glucose in Biological Matrixes:  Initialization Strategies and Effects of Spectral Resolution

1998 ◽  
Vol 70 (21) ◽  
pp. 4472-4479 ◽  
Author(s):  
Qing Ding ◽  
Gary W. Small ◽  
Mark A. Arnold
2019 ◽  
Vol 11 (36) ◽  
pp. 4593-4599
Author(s):  
Shaohui Yu ◽  
Jing Liu

A weighted clustering and pruning of wavelength variables-partial least squares (WCPV-PLS) method was proposed.


2013 ◽  
Vol 365-366 ◽  
pp. 737-740
Author(s):  
Li Jun Yao ◽  
Jie Mei Chen ◽  
Tao Pan

Near-infrared (NIR) spectroscopy combined with moving window partial least squares (MWPLS) method was successfully applied to the waveband selection for the rapid chemical-free determination of Zn2+ in soil. Based on randomness and similarity, an effective approach was performed to obtain objective and practical models. The optimal MWPLS waveband was 1136-1252 nm, and the corresponding optimal number of PLS factors was 6. The validation root mean square error (V-SEP) and validation correlation coefficients (V-RP) of prediction were 15.658 mg kg-1 and 0.925, respectively. The Zn2+ prediction values of the validation samples are close to the measured values. The results provided a reliable NIR model and can serve as valuable references for designing the dedicated spectroscopic instruments.


Sign in / Sign up

Export Citation Format

Share Document