Table 3.
Confusion matrix of grapevine varietal classification using support vector machines and an external validation of 40 samples. The diagonal of the matrix corresponds to the number of samples that were properly classified. The last column displays, for each variety, the correctly classified percentage (n = 40).
| Predicted Variety | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CS | CL | CR | WG | PX | PN | TE | TR | VO | VU | % | ||
| Actual variety | CS | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100.0 |
| CL | 0 | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100.0 | |
| CR | 0 | 0 | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 100.0 | |
| WG | 0 | 0 | 0 | 4 | 0 | 0 | 0 | 0 | 0 | 0 | 100.0 | |
| PX | 0 | 0 | 0 | 0 | 4 | 0 | 0 | 0 | 0 | 0 | 100.0 | |
| PN | 0 | 0 | 0 | 0 | 0 | 3 | 0 | 0 | 0 | 1 | 75.0 | |
| TE | 0 | 0 | 0 | 0 | 0 | 1 | 3 | 0 | 0 | 0 | 75.0 | |
| TR | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 0 | 0 | 100.0 | |
| VO | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 3 | 0 | 75.0 | |
| VU | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 4 | 100.0 | |
CS: Cabernet Sauvignon; CL: Caladoc; CR: Carmenere; WG: White Grenache; PX: Pedro Ximenez; PN: Pinot Noir; TE: Tempranillo; TR: Treixadura; VO: Viognier; VU: Viura.