Combination of machine learning and intelligent sensors in real-time quality control of alcoholic beverages

Abstract

Machine learning (ML) featured on its ability of learning and extracting features from a large set of data and automatically building statistical models. Through cooperation with intelligent sensors, which is designed to imitate human organs to analyze the sensory characteristics of foods, ML-based intelligent sensory systems such as electronic nose (E-nose) and electronic tongue (E-tongue) are developed for sensing applications in food industry. Consumption of alcohol beverages keep growing worldwide in recent years and fraudulent activities are stimulated due to the high price of alcoholic drinks, which motivates the application of intelligent sensory technology with high efficiency and accuracy for real-time quality control. Thus, this paper firstly summarizes the novel intelligent sensors that is suitable for sensory evaluation and the advanced ML algorithms used to create intelligent systems. Then the paper describes the mechanism of commercial ML-enabled intelligent devices and summarizes their practical sensing applications on the real-time quality control of a variety of alcoholic beverages, in term of detection of frauds and adulterations, aroma analysis, monitoring of the production process, and correlation with human sensory perception. Finally, the potential applications and future opportunities of ML-enabled intelligent sensor systems in the alcohol industry are discussed.

Qualitative and Quantitative Detection of CRISPR-Associated Cas Gene in Gene-Edited Foods

“…The results are shown in Figure 3D. All the 60 repeated qualitative PCR experiments yielded targeted amplification products, thus meeting the requirements for determining the qualitative detection limit of gene editing components [27]. Therefore, the detection limit of the method was taken as 0.1%.…”

Advancements in Improving Selectivity of Metal Oxide Semiconductor Gas Sensors Opening New Perspectives for Their Application in Food Industry

“…support vector machine (SVM), showed that the employed set of MOS gas detectors accurately identified pesticide residues in apple samples [193]. Gas detection is also essential in the production of various food products like coffee [194], beer [172], cheese [195], and wine [170,196], as well as baking processes [197], in which monitoring the presence of various gases allows the assessment of correctness of production processes and product quality. Experimental results show that the developed E-nose system using SnO2-based MOS gas sensors combined with six machine learning techniques (decision tree, random forest, XGBoost, SVM, convolutional neural network (CNN), and CNN+LSTM) achieves good performance in coffee aroma recognition [194].…”

Laser-Scribed Graphene-Based Electrochemical Sensors: A Review

“…LSG was applied to make up glucose biofuel cells, and the electrode was integrated with active catalytic nanomaterials to take advantage of the biochemical energy of glucose [17]. Applying a non-thermal laser scribe method on graphene oxide (GO)/silver nanowires (SNWs) thin film reduced it to rGO/SNW, resulting in 80% transparency and 70 Ω cm −1 sheet resistance that could be used as an optoelectronic device [18]. When Cu x O nanoparticle were immobilized on LSG a highly efficient and inexpensive electrocatalyst for hydrogen production resulted [19].…”

“…By coupling LSG electrochemical sensors to machine learning (ML) models, it is possible to achieve rapid data acquisition, and the intelligent and automated analysis of large amounts of data, with accurate results [18,19]. The association of mathematical and statistical methods to analyze, process, and extract crucial information from large complex datasets can be fundamental in complex cases [20].…”