Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data

Abstract

To date, many machine learning models have been used for peach maturity prediction using non-destructive data, but no performance comparison of the models on these datasets has been conducted. In this study, eight machine learning models were trained on a dataset containing data from 180 ‘Suncrest’ peaches. Before the models were trained, the dataset was subjected to dimensionality reduction using the least absolute shrinkage and selection operator (LASSO) regularization, and 8 input variables (out of 29) were chosen. At the same time, a subgroup consisting of the peach ground color measurements was singled out by dividing the set of variables into three subgroups and by using group LASSO regularization. This type of variable subgroup selection provided valuable information on the contribution of specific groups of peach traits to the maturity prediction. The area under the receiver operating characteristic curve (AUC) values of the selected models were compared, and the artificial neural network (ANN) model achieved the best performance, with an average AUC of 0.782. The second-best machine learning model was linear discriminant analysis with an AUC of 0.766, followed by logistic regression, gradient boosting machine, random forest, support vector machines, a classification and regression trees model, and k-nearest neighbors. Although the primary parameter used to determine the performance of the model was AUC, accuracy, F1 score, and kappa served as control parameters and ultimately confirmed the obtained results. By outperforming other models, ANN proved to be the most accurate model for peach maturity prediction on the given dataset.

Biochemical and molecular changes in peach fruit exposed to cold stress conditions

“…Recently, an artificial neural network (ANN) excelled above other models for the prediction of peach maturity based on a given dataset of non -destructive data, with the linear discriminant analysis ranking second, followed by logistic regression, gradient boosting machine, random forest, SVM, and k-nearest neighbours (Ljubobratović et al 2022). Xuan et al (2022) used hyperspectral imaging to evaluate inner and outer quality of a peach cultivar, by providing spectral as well as spatial data.…”

Shelf Life Prediction of Post-Harvested Pomegranate using Enhanced Deep Learning

“…The wings are used for improving the height, and the search agents make the spiral characteristics during the exploitation (attacking). This characteristic is described in the 3-D plane, and it is represented as: sin(k) Rad X '   (11) cos(k) Rad Y '   (12) k Rad Z '   (13) ) exp(v w r n   (14) where Rad is the radius of every turn of the spiral and k is the parameter which ranges between 0 and …”

“…In [Iorliam et al (2021)], ML based k-nearest neighbour (KNN), logistic regression (LR), and naïve Bayes (NB) approaches are introduced to predict the shelf life based on Firmness, PLW, acidic contents etc. In [Ljubobratović et al (2022)], ML based non-linear LR model is introduced to classify the shelf life prediction in orange fruit under different temperature conditions. In [Owoyemi et al (2022)], the ML-based approach is introduced to identify the maturity prediction in peach fruit.…”

A Bibliometric Analysis of Text Mining: Exploring the Use of Natural Language Processing in Social Media Research

“…Nowadays, voice assistance, personalized suggestions, medical diagnosing, treatment solutions, and many other essential fields use machine learning, a discipline that is constantly and rapidly expanding and developing. In this context, Callier [4] discusses the role of machine learning in evolutionary studies, Dejan et al [5] provide an assessment of various machine learning models for sensor-extracted data, and Telikani et al [6] provide a survey on evolutionary machine learning.…”