Classification of extra virgin olive oils via LIBS technique using ns laser pulses combined with machine learning algorithms

• Κύριος συγγραφέας: Γυφτοκώστας, Νικόλαος
• Άλλοι συγγραφείς: Gyftokostas, Nikolaos
• Γλώσσα: English
• Έκδοση: 2021
• Θέματα: LIBS; Laser induced breakdown spectroscopy; Machine learning; Olive oil classification

In the present work, 78 extra virgin olive oil samples (EVOOs) collected from different places of Greece, specifically from the areas of Crete, Peloponnese, Lesvos and Ionian islands, were studied by means of a laser-based technique called Laser-Induced Breakdown Spectroscopy (LIBS), assisted by well-known machine learning algorithms, such as, Principal Components Analysis (PCA), Linear Discriminant Analysis (LDA) and Support Vector Machines (SVMs), aiming to classify/discriminate them in terms of their geographical origin. The classification results obtained demonstrate nicely the potential of the combination of LIBS technique and machine learning algorithms, paving the way towards the utilization of LIBS technique for classification/discrimination of olive oils and eventually for other types of agricultural and edible products.