Using data mining techniques for classificatiom of essential oils according to yield

High yield is one of the main goals in the process of producing essential oils. The basic purpose of this study is the classification of essential oil based on known process parameters. A database was used consisting of data obtained from the experiments of of essential oil hydrodistillation from juniper berries. For each sample, in addition to the process parameters, a number of characteristics of the plant drug as raw material for the production of essential oil has been collected. Using 6 different algorithms for essential oil classification (Decision Tree, Random Forest, algorithm k Nearest Neighbors, Support Vector Machines, Neural Network and Naïve Bayes classifier), the yield of essential oils is classified in 4 classes. The most successful classification was obtained using the decision tree algorithm. These results can be used to support decision-making in the production of essential oils.