Title | ||
---|---|---|
Determination of Protein Content of Auricularia Auricula Using Spectroscopy and Least Squares-Support Vector Machine |
Abstract | ||
---|---|---|
Visible and near infrared (Vis/NIR) spectroscopy combined with calibration methods was investigated for the determination of protein content of auricularia auricula. The calibration set was consisted of 180 samples and the remaining 60 samples for the validation set. Different preprocessing methods were compared in partial least squares (PLS) models including Savitzky-Golay smoothing (SG), standard normal variate (SNV), the first and second derivative (1-Der and 2-Der), de-trending, and direct orthogonal signal correction (DOSC). The optimal PLS model was achieved by DOSC-PLS with determination coefficient R2 = 0.9533 and root mean squares error of prediction RMSEP = 0.1884. Simultaneously, the scores of PLS latent variables were employed as the inputs of least squares-support vector machine (LS-SVM).The optimal prediction results were R2 = 0.9830 and RMSEP = 0.1146 which was better than DOSC-PLS model. The results indicated that Vis/NIR spectroscopy combined with LS-SVM could be utilized as an efficient way for the determination of protein content of auricularia auricula. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICNC.2009.495 | ICNC (1) |
Keywords | Field | DocType |
savitzky-golay smoothing,nir spectroscopy,dosc-pls model,calibration,near infrared spectroscopy,protein content,auricularia auricula protein content determination,mean squares error,visible spectroscopy,proteins,direct orthogonal signal correction,auricularia auricula,least squares support vector machine,pls latent variable,vis/nlr spectroscopy,least squares approximations,optimal pls model,standard normal variate,determination coefficient r2,partial least squares model,calibration set,calibration method,root mean squares error,infrared spectroscopy,biocomputing,support vector machines,mean square error methods,squares-support vector machine,kernel,computational modeling,near infrared,spectroscopy,predictive models,latent variable | Biological system,Computer science,Partial least squares regression,Artificial intelligence,Second derivative,Least squares support vector machine,Support vector machine,Near-infrared spectroscopy,Smoothing,Spectroscopy,Statistics,Machine learning,Calibration | Conference |
Volume | ISBN | Citations |
1 | 978-0-7695-3736-8 | 0 |
PageRank | References | Authors |
0.34 | 3 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Fei Liu | 1 | 0 | 0.34 |
Guangming Sun | 2 | 0 | 0.34 |
Yong He | 3 | 78 | 12.64 |