Journal of Analytical Methods in Chemistry / 2019 / Article / Tab 5 / Research Article
Simultaneous Recognition of Species, Quality Grades, and Multivariate Calibration of Antioxidant Activities for 12 Famous Green Teas Using Mid- and Near-Infrared Spectroscopy Coupled with Chemometrics Table 5 The predictions of ABTS scavenging activities by OSWLS-SVM.
Prediction samples Actual ABTS radical scavenging activity (TEAC) Recoveries (%) MIR NIR Fusion data MIR NIR Fusion data 1 4.44 4.44 4.44 97.4 99.4 100.0 2 4.46 4.44 4.44 100.0 99.8 100.0 3 4.44 4.46 4.43 96.1 100.5 100.4 4 4.46 4.44 4.44 98.2 99.6 100.7 5 4.11 4.44 4.11 100.5 99.2 100.2 6 4.11 4.46 4.11 100.0 99.5 100.0 7 4.12 4.11 4.14 100.0 100.7 100.0 8 4.14 4.14 4.14 100.0 99.8 99.3 9 4.11 4.11 4.11 100.5 99.8 100.3 10 4.12 4.11 4.12 100.2 99.4 100.0 11 4.14 4.03 4.14 100.0 100.2 100.0 12 4.04 4.04 4.04 100.0 99.7 100.0 13 4.04 4.04 4.04 100.0 100.3 100.1 14 4.04 4.04 4.04 100.0 99.7 100.2 15 4.04 4.04 4.04 97.8 97.0 100.0 16 4.21 4.21 4.21 100.1 101.2 100.3 17 4.22 4.22 4.22 100.0 100.0 100.0 18 4.23 4.23 4.23 99.8 100.4 99.9 19 4.22 4.22 4.22 100.2 100.4 100.1 20 4.23 4.23 4.23 99.5 98.8 100.0 21 4.33 4.33 4.33 100.0 101.8 100.3 22 4.35 4.35 4.35 99.7 100.7 99.9 23 4.36 4.33 4.36 99.5 100.4 100.0 24 4.33 4.35 4.33 100.0 99.2 100.0 25 4.36 4.36 4.36 100.0 100.5 100.0 26 4.27 4.28 4.27 100.4 100.4 100.0 27 4.28 4.29 4.28 100.0 99.8 98.4 28 4.30 4.30 4.30 100.0 99.4 100.0 29 4.27 4.27 4.27 100.0 100.3 100.0 30 4.29 4.27 4.29 99.8 100.4 99.7 Correlation coefficient 1.0000 0.9990 1.0000 Regression equation y = 1.0000x + 3.6935 × 10−6 y = 0.9860x + 0.0565y = 1.0000x + 2.7420 × 10−6 Average recoveries (%) 99.7 ± 1.0 99.9 ± 0.8 100.0 ± 0.4 RMSECa 1.1315 × 10−6 0.0394 7.6170 × 10−7 RMSEPb 0.0447 0.0347 0.0155 T(t -test) = 2.045 0.4615 < 0.0634 < 0.0054 <
a RMSEC denotes calibration root-mean-squared error; b RMSEP denotes prediction root-mean-squared error.