Application of near infrared spectroscopy combined with SVR algorithm in rapid detection of cAMP content in red jujube

In order to further improve the performance of the near-infrared (NIR) spectroscopy quantitative model for detecting cyclic adenosine monophosphate (cAMP) content in red jujube, in this paper, support vector regression (SVR) is used for spectral analysis and compared with partial least squares (PLS) model results. The results show that for PLS model, correction coefficient (R-c(2)), correction set root mean square error of calibration (RMSEC), prediction coefficient (R-p(2)) and prediction set root mean square error of prediction (RMSEP) are 0.9076, 25.2625, 0.8323 and 29.0407, respectively. The performance of the SVR model is much better, and R-c(2), RMSEC, R-p(2) and RMSEP are0.9850, 11.1233, 0.9388 and 13.0739, respectively. The research indicates that the SVR model can greatly improve the predictive performance and stability of the jujube cAMP quantitative model.