Random Forest, Artificial Neural Network, and Support Vector Machine Models for Honey Classification
- Cecilia Martinez-Castillo1, 2, Gonzalo Astray1, *, Juan Carlos Mejuto1, Jesus Simal-Gandara2, *1Department of Physical Chemistry, Faculty of Food Science and Technology, University of Vigo - Ourense Campus, Ourense E32004, Spain2Nutrition and Bromatology Group, Department of Analytical and Food Chemistry, Faculty of Food Science and Technology, University of Vigo - Ourense Campus, Ourense E32004, Spain
- Corresponding Author
- Gonzalo Astray
- https://doi.org/10.2991/efood.k.191004.001How to use a DOI?
- Food authenticity, honey, Galician honeys, classification models
Different separated protein fractions by the electrophoretic method in polyacrylamide gel were used to classify two different types of honeys, Galician honeys and commercial honeys produced and packaged outside of Galicia. Random forest, artificial neural network, and support vector machine models were tested to differentiate Galician honeys and other commercial honeys produced and packaged outside of Galicia. The results obtained for the best random forest model allowed us to determine the origin of honeys with an accuracy of 95.2%. The random forest model, and the other developed models, could be improved with the inclusion of new data from different commercial honeys.
- © 2019 International Association of Dietetic Nutrition and Safety. Publishing services by Atlantis Press International B.V.
- Open Access
- This is an open access article distributed under the CC BY-NC 4.0 license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - JOUR AU - Cecilia Martinez-Castillo AU - Gonzalo Astray AU - Juan Carlos Mejuto AU - Jesus Simal-Gandara PY - 2019 DA - 2019/10 TI - Random Forest, Artificial Neural Network, and Support Vector Machine Models for Honey Classification JO - eFood SN - 2666-3066 UR - https://doi.org/10.2991/efood.k.191004.001 DO - https://doi.org/10.2991/efood.k.191004.001 ID - Martinez-Castillo2019 ER -