Confusion matrix from the execution with the best score (ANN, SNV+D, D2W5 and parameter set 10) with an overall correctly classified value of 87.25% (20 leaves per variety).

Each row represents the actual variety and in which one was classified. Bolded values (diagonal of the matrix) are the number of samples properly classified. The last column shows the correctly classified percentage for each variety.ANN: Artificial Neural Network; SNV+D: Standard Normal Variate followed by De-trending; D2W5: Second-degree derivative and window size 5 Savitzky-Golay filter.Ve: Verdejo; M: Malvasia; V: Viura; A: Albariño; T: Treixadura; G: Godello; WG: White Grenache; WT: White Tempranillo; PX: Pedro Ximénez; Vi: Viognier; CF: Cabernet Franc; Gr: Grenache; CS: Cabernet Sauvignon; C: Carmenere; S: Syrah; Te: Tempranillo; PN: Pinot Noir; Ca: Caladoc; Ma: Marselan; TN: Touriga Nacional.Confusion matrix from the execution with the best score (ANN, SNV+D, D2W5 and parameter set 10) with an overall correctly classified value of 87.25% (20 leaves per variety).