Skip to main content
. 2021 Jun 7;21(11):3923. doi: 10.3390/s21113923

Table 1.

Testing results from six different classifier models for wine samples.

Classifier Type Sensitivity Specificity PPV NPV Accuracy
Linear
discriminant
Zinfandel 1 1 1 1 1
Cabernet sauvignon 1 1 1 1 1
Pinot noir 1 0.8889 1 0.9643 0.9722
Merlot 0.963 1 0.9 1 0.9722
Quadratic
discriminant
Zinfandel 1 1 1 1 1
Cabernet sauvignon 1 1 1 1 1
Pinot noir 1 0.5 1 0.8571 0.875
Merlot 0.833 1 0.6667 1 0.875
Linear
SVM
Zinfandel 1 1 1 1 1
Cabernet sauvignon 1 1 1 1 1
Pinot noir 1 0.944 1 0.9818 0.9861
Merlot 0.9815 1 0.9474 1 0.9861
Quadratic
SVM
Zinfandel 1 1 1 1 1
Cabernet sauvignon 1 1 1 1 1
Pinot noir 1 0.944 1 0.9818 0.9861
Merlot 0.9815 1 0.9474 1 0.9861
Bayes
Gaussian
Zinfandel 1 1 1 1 1
Cabernet sauvignon 1 0.5 1 0.8571 0.875
Pinot noir 1 0.833 1 0.9474 0.9583
Merlot 0.7778 1 0.6 1 0.8333
KNN
fine
Zinfandel 1 1 1 1 1
Cabernet sauvignon 1 0.5 1 0.8571 0.875
Pinot noir 1 0.9444 1 0.9818 0.9861
Merlot 0.8148 1 0.6429 1 0.8611