2.4.1. SIMCA

Classification analyses of maple syrups were performed using a supervised pattern recognition classification method SIMCA, which uses the previous understanding of the category membership of samples to classify new unrevealed samples in one of the known classes based on the pattern of measurements [23]. The cross-validation (leave-out-out) was used to assess the performance of the training model by analyzing the misclassification and generalization error [24]. The performance of the SIMCA model was also assessed with class projections, discriminating power, misclassification, and interclass distances (ICD), which interpret the quantitative similarity or dissimilarity of different classes and are widely accepted as samples that can be well differentiated when ICD >3 [24].
