Tomato storage quality predicting method based on portable electronic nose system combined with WOA-SVM model

In this paper, a portable electronic nose (e-nose) system, combined with the support vector machine model characterized by whale optimization algorithm (WOA-SVM), was developed for tomato storage quality prediction. K-nearest neighbor (KNN) model, decision tree (DT) model, support vector machine (SVM) model and support vector machine combined with loop optimization algorithm (LOA-SVM) model were constructed to provide a comparison for the WOA-SVM model. E-nose responses at different storage times were collected for 21 days. Results demonstrated that the prediction accuracy of the WOA-SVM model reached 99.81%, which was high than other models. This method presented some advantages including rapid analysis, good repeatability and high accuracy.