Journal of Analytical Methods in Chemistry / 2015 / Article / Tab 2 / Research Article
Determination of Cefoperazone Sodium in Presence of Related Impurities by Linear Support Vector Regression and Partial Least Squares Chemometric Models Table 2 Assay results for prediction of training set (autoprediction) and independent test set of CEF by PLSR and linear SVR chemometric models.
Training set PLSR Linear SVR Taken (µ g mL−1 ) Found (µ g mL−1 ) %Found (µ g mL−1 ) %18 18.577 103.21 18.629 103.50 18 18.029 100.16 18.136 100.76 20 19.850 99.25 19.920 99.60 20 19.889 99.44 19.980 99.90 26 26.114 100.44 26.110 100.42 20 19.899 99.49 19.971 99.85 18 17.897 99.43 17.980 99.89 24 23.737 98.90 23.754 98.97 24 23.997 99.99 23.980 99.92 20 19.727 98.63 19.775 98.87 24 24.005 100.02 23.980 99.92 18 18.089 100.50 18.116 100.65 26 26.236 100.91 26.143 100.55 26 25.919 99.69 25.980 99.92 24 24.118 100.49 24.181 100.76 26 25.917 99.68 25.980 99.92 Mean (%) 100.01 100.21 SD 1.050 1.036 RMSEC 0.2306 0.1960 Test set PLSR Linear SVR Taken (µ g mL−1 ) Found (µ g mL−1 ) %Found (µ g mL−1 ) %25 25.443 101.77 25.320 101.28 19 19.069 100.36 19.067 100.35 22 22.357 101.62 22.278 101.27 21 21.403 101.92 21.330 101.57 23 23.515 102.24 23.397 101.73 25 25.630 102.52 25.440 101.76 19 19.294 101.55 19.238 101.25 22 22.432 101.96 22.296 101.35 21 21.603 102.87 21.493 102.35 Mean (%) 101.87 101.43 SD 0.7082 0.5359 RMSEP 0.4457 0.3386