Skip to main content

Table 3 The performance of four kinds of regression models

From: Determination of manganese content in cottonseed meal using near-infrared spectrometry and multivariate calibration

Models Varialbes Calibration Validation RPD RMSECV
Num R 2 c RMSEC Num R 2 v RMSEP
PLS 700 216 0.894 7 2.598 3 72 0.926 9 2.438 7 3.574 6 3.171 9
LSSVM 700 216 0.984 9 1.001 7 72 0.944 1 2.108 4 4.134 7 2.491 0
MCUVE-LSSVM 233 216 0.976 3 1.242 0 72 0.948 6 2.010 8 4.335 4 2.309 1
MCUVE-SPA-LSSVM 49 216 0.972 9 1.330 8 72 0.949 3 1.994 6 4.370 5 2.101 3
  1. Num Number of samples used to perform the calibration and validation, R2c Coefficient of determination in calibration, R2v Coefficient of determination in calibration, RMSEC The root mean square error of calibration, RMSEP The root mean square error of prediction, RPD The residual predictive deviation, RMSECV The root mean square error of cross-valid