Electronic Nose Coupled with Linear and Nonlinear Supervised Learning Methods for Rapid Discriminating Quality Grades of Superior Java Cocoa Beans

Abstract

An electronic nose (E-nose), comprising eight metal oxide semiconductor (MOS) gas sensors and a moisture-temperature sensor, was used for classifying three quality grades of superior java cocoa beans, namely fine cocoa dark bean < 20%, fine cocoa dark bean > 60%, and bulk cocoa bean that is a harder task compared to the discrimination of high versus low-quality cocoa beans. The E-nose signals were pre-processed using the maximum value method. The capability for discriminating the quality grade of the cocoa beans was checked by applying multivariate statistical tools, namely, linear discriminant analysis (LDA), support vector machine (SVM) and artificial neural networks (ANN). For this, the experimental dataset was split into two subsets, one for training (i.e., establishing the classification models) and the other for external-validation purposes. Furthermore, hyperparameter optimization and K-fold cross-validation variant were implemented during the model training procedure to select the best classification models and to avoid over-fitting issues. The best predictive classification performance was obtained with the E-nose-MLP-ANN procedure, which allowed 99% of correct classifications (overall accuracy) for the training dataset and 95% of correct classifications (overall accuracy) for the external-validation dataset. The satisfactory results clearly demonstrated that the E-nose could be applied as a quality control tool in the cocoa industry, requiring minimum and simple sample preparation.

Intelligent Mobile Electronic Nose System Comprising a Hybrid Polymer-Functionalized Quartz Crystal Microbalance Sensor Array

“…For machine learning and pattern-classification applications, the LDA model was normally employed as a technique for dimensionality reduction and classification at the preprocessing stage. In addition to the LDA model, an SVM algorithm was also applied to QCM e-nose data processing as a comparison, in which this machine learning technique was suitable for handling linear and nonlinear data with complex pattern recognition problems . SVM uses a quadratic hyperplane optimization to discriminate the classes and a kernel function to optimize its performance based on a linear, polynomial, or radial basis function.…”

“…In addition to the LDA model, an SVM algorithm was also applied to QCM e-nose data processing as a comparison, in which this machine learning technique was suitable for handling linear and nonlinear data with complex pattern recognition problems. 10 SVM uses a quadratic hyperplane optimization to discriminate the classes and a kernel function to optimize its performance based on a linear, polynomial, or radial basis function. A grid search procedure with a 10-fold cross-validation method was applied for hyperparameter tuning.…”

“…Electronic nose (e-nose) technology has been developed by scientists worldwide and greatly used in many application fields by mimicking and advancing the olfactory function of human beings, e.g., in medical care for noninvasive early disease diagnosis, − food industry for product quality assurance, − environmental monitoring for continuously online/in situ hazardous gas detection, and agriculture for plant protection − via monitoring and classification of volatile organic compounds (VOCs). An e-nose system consists normally of some or many gas sensors that are assembled into an array targeting different analytes, a so-called sensor array.…”

Optimizing Machine Learning Parameters for Classifying the Sweetness of Pineapple Aroma Using Electronic Nose

“…Previous studies have also carried out the classification process using E-nose for Java cocoa fruit [7]. The best predictive classification process is obtained by the E-nose -MLP-ANN procedure.…”