Πρόβλεψη σοδειάς βιομηχανικής τομάτας με μεθόδους μηχανικής μάθησης
Industrial tomato yield prediction using machine learning

View/ Open
Keywords
Επιστήμη Δεδομένων ; Ridge ; Απόδοση ; Μηχανική μάθηση ; Ντομάτα ; Μοντέλα πρόβλεψης ; Επιστήμη δεδομένωνAbstract
Prediction models are extensively used due to improvements in computer science. The availability of
data combined with machine learning algorithms allows the extraction of feature importance of the
variables and accurate production forecasting. Cultivation data from industrial tomatoes, tomatoes are
one of the most widely produced and consumed vegetables in the world, is used to accurately predict
production. In this thesis, a model has been developed to predict the production of industrial tomato
crops based on previously collected data. Data were collected from different fields in different regions
of the Peloponnese in Greece, over 3 growing seasons. In order to find the optimal algorithm for this
data set, many different algorithms were tested on different measurements. Ridge was used in order to
develop the prediction model.