Machine Learning in Sensory Analysis of Mead. Case Study: Ensembles of Classifiers

This research focuses on the application of machine learning to the sensory analysis of mead, which opens up new possibilities in its classification and understanding. The aim was to use machine learning algorithms to classify mead types based on their sensory analysis. Machine learning algorithms such as Random Forest (RF), Adaptive Boosting (AdaBoost), Bootstrap Aggregating (Bagging), K-Nearest Neighbours (KNN) and Decision Tree (DT) were used in the analysis of chemical and sensory datasets. The Random Forest and K-Nearest Neighbours (KNN) algorithms were found to be the most effective in mead recognition, obtaining the highest scores. In contrast, the AdaBoost algorithm consistently produced the lowest accuracy results. Nonetheless, the Decision Tree algorithm achieved the highest accuracy value (0.909), indicating its potential in accurate classification based on aroma characteristics. The results suggest that the choice of an appropriate classification model can significantly affect the performance of the mead identification process in practical applications. Machine learning offers new opportunities in optimising mead production processes. The application of machine learning in the sensory analysis of mead is important for accurate classification, a better understanding of the factors affecting quality and the optimization of the production processes of this beverage, contributing to the development of interdisciplinary food research.