%0 Journal Article %J Food Control, Elsevier Science %D 2022 %T Honey discrimination based on the bee feeding by Laser Induced Breakdown Spectroscopy %A Dimitris Stefas %A Nikolaos Gyftokostas %A Panagiotis Kourelias %A Eleni Nanou %A Chrysoula Tananaki %A Dimitrios Kanelis %A Vasileios Liolios %A Vasileios Kokkinos %A Christos Bouras %A Stelios Couris %X In the present work, the effects of artificial feeding of bees on the honey are investigated by employing for the first time, Laser Induced Breakdown Spectroscopy (LIBS) by analyzing the emission spectral characteristics of the plasma created on the surface of honey samples. Correlation plots indicating the importance of spectral lines of elements as e.g., Calcium (Ca), Magnesium (Mg), Sodium (Na) and Potassium (K) are constructed, while machine learning algorithms based on Linear Discriminant Analysis (LDA) and Random Forest Classifiers (RFC) are employed to classify the honey samples in terms of the bee food used. The constructed machine learning models were validated by both cross-validation and external validation, while the obtained accuracies exceeded 90% of correct classification, indicating the potential of LIBS technique for honey discrimination. The obtained results by LIBS were also validated by HPLC-RID, which is the standard technique used for the analysis of the main honey sugars. %B Food Control, Elsevier Science %G eng %0 Journal Article %J Open Access Journal, MDPI, Special Issue Characterization of Olive Products from Greece %D 2021 %T Classification of Greek olive from different regions by machine learning - aided Laser - Induced Breakdown Spectroscopy and Absorption Spectroscopy %A Nikolaos Gyftokostas %A Eleni Nanou %A Dimitrios Stefas %A Vasileios Kokkinos %A Christos Bouras %A Stelios Couris %X In the present work, the emission and the absorption spectra of numerous Greek olive oil samples and mixtures of them, obtained by two spectroscopic techniques, namely Laser-Induced Breakdown Spectroscopy (LIBS) and Absorption Spectroscopy, and aided by machine learning algorithms, were employed for the discrimination/classification of olive oils regarding their geographical origin. Both emission and absorption spectra were initially preprocessed by means of Principal Component Analysis (PCA) and were subsequently used for the construction of predictive models, employing Linear Discriminant Analysis (LDA) and Support Vector Machines (SVM). All data analysis methodologies were validated by both “k-fold” cross-validation and external validation methods. In all cases, very high classification accuracies were found, up to 100%. The present results demonstrate the advantages of machine learning implementation for improving the capabilities of these spectroscopic techniques as tools for efficient olive oil quality monitoring and control. %B Open Access Journal, MDPI, Special Issue Characterization of Olive Products from Greece %V 26 %G eng %N 5 %0 Journal Article %J Food Control, Elsevier Science %D 2021 %T Discrimination of olive oils based on the olive cultivar origin by machine learning employing the fusion of emission and absorption spectroscopic data %A Christos Bouras %A Dimitrios Stefas %A Nikolaos Gyftokostas %A Panagiotis Kourelias %A Eleni Nanou %A Vasileios Kokkinos %A Stelios Couris %X In this work Laser-Induced Breakdown Spectroscopy (LIBS) and absorption spectroscopy aided by machine learning are employed for discriminating some extra virgin Greek olive oils of different olive cultivars for the first time. LIBS and absorption spectra of extra virgin olive oils belonging to Kolovi and Koroneiki cultivars, as well as mixtures of them, were collected, analyzed, and used to develop classification schemes employing Linear Discriminant Analysis and Gradient Boosting, the latter allowing the determination of the most important spectral features. Both algorithms were found to provide efficient classification of the olive oil spectra with accuracies exceeding 90%. Furthermore, for the first time, the emission spectra of LIBS were fused with the absorption spectra to create predictive models and their accuracies were found to be significantly improved. This work demonstrates the enhanced capabilities of LIBS and absorption spectroscopy and the potential of their combination for olive oil quality monitoring and control. %B Food Control, Elsevier Science %V 130 %P 1-8 %G eng %0 Journal Article %J Scientific Reports, Nature Research Journals %D 2021 %T Laser-Induced Breakdown Spectroscopy coupled with machine learning as a tool for olive oil authenticity and geographic discrimination %A Nikolaos Gyftokostas %A Dimitrios Stefas %A Vasileios Kokkinos %A Christos Bouras %A Stelios Couris %X Olive oil is a basic element of the Mediterranean diet and a key product for the economies of the Mediterranean countries. Thus, there is an added incentive in the olive oil business for fraud through practices like adulteration and mislabeling. In the present work, Laser Induced Breakdown Spectroscopy (LIBS) assisted by machine learning is used for the classification of 139 virgin olive oils in terms of their geographical origin. The LIBS spectra of these olive oil samples were used to train different machine learning algorithms, namely LDA, ERTC, RFC, XGBoost, and to assess their classification performance. In addition, the variable importance of the spectral features was calculated, for the identification of the most important ones for the classification performance and to reduce their number for the algorithmic training. The algorithmic training was evaluated and tested by means of classification reports, confusion matrices and by external validation procedure as well. The present results demonstrate that machine learning aided LIBS can be a powerful and efficient tool for the rapid authentication of the geographic origin of virgin olive oil. %B Scientific Reports, Nature Research Journals %V 11 %G eng %0 Journal Article %J Applied Sciences, MPDI, Special Issue Chemical Composition, Properties and Applications of Honey %D 2021 %T A Laser‐Based Method for the Detection of Honey Adulteration %A Dimitrios Stefas %A Nikolaos Gyftokostas %A Panagiotis Kourelias %A Eleni Nanou %A Vasileios Kokkinos %A Christos Bouras %A Stelios Couris %X In the present work, laser‐induced breakdown spectroscopy, aided by some machine learning algorithms (i.e., linear discriminant analysis (LDA) and extremely randomized trees (ERT)), is used for the detection of honey adulteration with glucose syrup. In addition, it is shown that instead of the entire LIBS spectrum, the spectral lines of inorganic ingredients of honey (i.e., calcium, sodium, and potassium) can be also used for the detection of adulteration providing efficient discrimination. The constructed predictive models attained high classification accuracies exceeding 90% correct classification. %B Applied Sciences, MPDI, Special Issue Chemical Composition, Properties and Applications of Honey %G eng