Cross Validation Mean squared error = 0.324891.
Leave-one-out cross validation, mean enrichment, and ROC curves are the main evaluation techniques.
We employed leave-one-out cross validation, mean enrichment, tenfold cross validation, and ROC curves to evaluate our proposed method and other existing methods.
All models were applied to the same set of 1807 images, using 10-fold cross validation; mean values across five runs are shown.
The standard errors of cross validation (SECV) indicate a high degree of precision in the estimates of the concentration of both constituents (SECV = 0.78 and 2.37 for FN and ADF respectively), with highly significant coefficients of determination of the cross validation (mean R = 0.90 and 0.90 for FN and ADF respectively).
Tenfold cross-validation means that the available examples are partitioned into ten disjoint subsets.
Nested cross-validation means that there is an outer cross-validation loop for model assessment and an inner loop for model selection.
Briefly, to perform a 10-fold cross-validation means to randomly divide the dataset into ten portions of the same size.
This performance was obtained under various cross-validation settings (2-fold, 5-fold, 10-fold, leave-one-out - where n-fold cross-validation means that (n-1)/n of the dataset is used to learn, and the remaining 1/n for prediction - this is repeated n times, and the average performance is reported).
These parameters are the correlation coefficient R 2, predictive squared correlation coefficient q 2 and (q_{cv}^{2}) obtained from cross-validation, mean absolute error (MAE) and root mean square error (RMSE), which represent the goodness-of-fit, robustness and predictive behavior of the model, respectively [42].
The results show that the framework was able to help produce lower estimation errors than previous research, and the model built by the Support Vector Regression algorithm on the features selected by Elastic Net has the least cross-validation mean squared error.
