Using UV–Vis spectroscopy for simultaneous geographical and varietal classification of tea infusions simulating a home-made tea cup

Fingerprinting black tea: When spectroscopy meets machine learning a novel workflow for geographical origin identification

Food fraud, along with many challenges to the integrity and sustainability, threatens the prosperity of businesses and society as a whole. Tea is the second most commonly consumed non-alcoholic beverage globally. Challenges to tea authenticity require the development of highly efficient and rapid solutions to improve supply chain transparency. This study has produced an innovative workflow for black tea geographical indications (GI) discrimination based on non-targeted spectroscopic fingerprinting techniques. A total of 360 samples originating from nine GI regions worldwide were analysed by Fourier Transform Infrared (FTIR) and Near Infrared spectroscopy. Machine learning algorithms (k-nearest neighbours and support vector machine models) applied to the test data greatly improved the GI identification achieving 100% accuracy using FTIR. This workflow will provide a low-cost and user-friendly solution for on-site and real-time determination of black tea geographical origin along supply chains.

The Authentication of Gayo Arabica Green Coffee Beans with Different Cherry Processing Methods Using Portable LED-Based Fluorescence Spectroscopy and Chemometrics Analysis

Aceh is an important region for the production of high-quality Gayo arabica coffee in Indonesia. In this area, several coffee cherry processing methods are well implemented including the honey process (HP), wine process (WP), and natural process (NP). The most significant difference between the three coffee cherry processing methods is the fermentation process: HP is a process of pulped coffee bean fermentation, WP is coffee cherry fermentation, and NP is no fermentation. It is well known that the WP green coffee beans are better in quality and are sold at higher prices compared with the HP and NP green coffee beans. In this present study, we evaluated the utilization of fluorescence information to discriminate Gayo arabica green coffee beans from different cherry processing methods using portable fluorescence spectroscopy and chemometrics analysis. A total of 300 samples were used (n = 100 for HP, WP, and NP, respectively). Each sample consisted of three selected non-defective green coffee beans. Fluorescence spectral data from 348.5 nm to 866.5 nm were obtained by exciting the intact green coffee beans using a portable spectrometer equipped with four 365 nm LED lamps. The result showed that the fermented green coffee beans (HP and WP) were closely mapped and mostly clustered on the left side of PC1, with negative scores. The non-fermented (NP) green coffee beans were clustered mostly on the right of PC1 with positive scores. The results of the classification using partial least squares-discriminant analysis (PLS-DA), linear discriminant analysis (LDA), and principal component analysis-linear discriminant analysis (PCA-LDA) are acceptable, with an accuracy of more than 80% reported. The highest accuracy of prediction of 96.67% was obtained by using the PCA-LDA model. Our recent results show the potential application of portable fluorescence spectroscopy using LED lamps to classify and authenticate the Gayo arabica green coffee beans according to their different cherry processing methods. This innovative method is more affordable and could be easy to implement (in terms of both affordability and practicability) in the coffee industry in Indonesia.

Comparative Analysis of Hot and Cold Brews from Single-Estate Teas (Camellia sinensis) Grown across Europe: An Emerging Specialty Product

Tea is grown around the world under extremely diverse geographic and climatic conditions, namely, in China, India, the Far East and Africa. However, recently, growing tea also appears to be feasible in many regions of Europe, from where high-quality, chemical-free, organic, single-estate teas have been obtained. Hence, the aim of this study was to characterize the health-promoting properties in terms of the antioxidant capacity of traditional hot brews as well as cold brews of black, green and white teas produced across the European territory using a panel of antioxidant assays. Total polyphenol/flavonoid contents and metal chelating activity were also determined. For differentiating the characteristics of the different tea brews, ultraviolet-visible (UV-Vis) spectroscopy and ultra-high performance liquid chromatography coupled with high-resolution mass spectrometry were employed. Overall, our findings demonstrate for the first time that teas grown in Europe are good quality teas that are endowed with levels of health-promoting polyphenols and flavonoids and that have an antioxidant capacity similar to those grown in other parts of the world. This research is a vital contribution to the characterization of European teas, providing essential and important information for both European tea growers and consumers, and could be of guidance and support for the selection of teas grown in the old continent, along with having the best brewing conditions for maximizing the health benefits of tea.

Tea authentication and determination of chemical constituents using digital image-based fingerprint signatures and chemometrics

Tea (Camellia sinensis) fraud has been frequently identified and involves tampering with the labelling of inferior products or without geographical origin certification and even mixing them with superior quality teas to mask an adulteration. Consequently, economic losses and health damage to consumers are observed. Thus, a Chemometrics-assisted Color Histogram-based Analytical System (CACHAS) was employed a simple, cost-effective, reliable, and green analytical tool to screen the quality of teas. Data-Driven Soft Independent Modeling of Class Analogy was used to authenticate their geographical origin and category simultaneously, recognizing correctly all Argentinean and Sri Lankan black teas and Argentinean green teas. For the determination of moisture, total polyphenols, and caffeine, Partial Least Squares obtained satisfactory predictive abilities, with values of root mean squared error of prediction (RMSEP) of 0.50, 0.788, and 0.25 mg kg^-1, r_pred of 0.81, 0.902, and 0.81, and relative error of prediction (REP) of 6.38, 9.031, and 14.58%., respectively. CACHAS proved to be a good alternative tool for environmentally-friendly non-destructive chemical analysis.

Targeted HPLC-UV Polyphenolic Profiling to Detect and Quantify Adulterated Tea Samples by Chemometrics

Tea can be found among the most widely consumed beverages, but it is also highly susceptible to fraudulent practices of adulteration with other plants such as chicory to obtain an illicit economic gain. Simple, feasible and cheap analytical methods to assess tea authentication are therefore required. In the present contribution, a targeted HPLC-UV method for polyphenolic profiling, monitoring 17 polyphenolic and phenolic acids typically described in tea, was proposed to classify and authenticate tea samples versus chicory. For that purpose, the obtained HPLC-UV polyphenolic profiles (based on the peak areas at three different acquisition wavelengths) were employed as sample chemical descriptors for principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA) studies. Overall, PLS-DA demonstrated good sample grouping and discrimination of chicory against any tea variety, but also among the five different tea varieties under study, with classification errors below 8% and 10.5% for calibration and cross-validation, respectively. In addition, the potential use of polyphenolic profiles as chemical descriptors to detect and quantify frauds was evaluated by studying the adulteration of each tea variety with chicory, as well as the adulteration of red tea extracts with oolong tea extracts. Excellent results were obtained in all cases, with calibration, cross-validation, and prediction errors below 2.0%, 4.2%, and 3.9%, respectively, when using chicory as an adulterant, clearly improving on previously reported results when using non-targeted HPLC-UV fingerprinting methodologies.

NIR Spectrometric Approach for Geographical Origin Identification and Taste Related Compounds Content Prediction of Lushan Yunwu Tea

Lushan Yunwu Tea is one of a unique Chinese tea series, and total polyphenols (TP), free amino acids (FAA), and polyphenols-to-amino acids ratio models (TP/FAA) represent its most important taste-related indicators. In this work, a feasibility study was proposed to simultaneously predict the authenticity identification and taste-related indicators of Lushan Yunwu tea, using near-infrared spectroscopy combined with multivariate analysis. Different waveband selections and spectral pre-processing methods were compared during the discriminant analysis (DA) and partial least squares (PLS) model-building process. The DA model achieved optimal performance in distinguishing Lushan Yunwu tea from other non-Lushan Yunwu teas, with a correct classification rate of up to 100%. The synergy interval partial least squares (siPLS) and backward interval partial least squares (biPLS) algorithms showed considerable advantages in improving the prediction performance of TP, FAA, and TP/FAA. The siPLS algorithms achieved the best prediction results for TP (R_P = 0.9407, RPD = 3.00), FAA (R_P = 0.9110, RPD = 2.21) and TP/FAA (R_P = 0.9377, RPD = 2.90). These results indicated that NIR spectroscopy was a useful and low-cost tool by which to offer definitive quantitative and qualitative analysis for Lushan Yunwu tea.

High-Throughput Flow Injection Analysis-Mass Spectrometry (FIA-MS) Fingerprinting for the Authentication of Tea Application to the Detection of Teas Adulterated with Chicory

Tea is a broadly consumed beverage worldwide that is susceptible to fraudulent practices, including its adulteration with other plants such as chicory extracts. In the present work, a non-targeted high-throughput flow injection analysis-mass spectrometry (FIA-MS) fingerprinting methodology was employed to characterize and classify different varieties of tea (black, green, red, oolong, and white) and chicory extracts by principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA). Detection and quantitation of frauds in black and green tea extracts adulterated with chicory were also evaluated as proofs of concept using partial least squares (PLS) regression. Overall, PLS-DA showed that FIA-MS fingerprints in both negative and positive ionization modes were excellent sample chemical descriptors to discriminate tea samples from chicory independently of the tea product variety as well as to classify and discriminate among some of the analyzed tea groups. The classification rate was 100% in all the paired cases-i.e., each tea product variety versus chicory-by PLS-DA calibration and prediction models showing their capability to assess tea authentication. The results obtained for chicory adulteration detection and quantitation using PLS were satisfactory in the two adulteration cases evaluated (green and black teas adulterated with chicory), with calibration, cross-validation, and prediction errors below 5.8%, 8.5%, and 16.4%, respectively. Thus, the non-targeted FIA-MS fingerprinting methodology demonstrated to be a high-throughput, cost-effective, simple, and reliable approach to assess tea authentication issues.

Starch-Based Hydrogel Nanoparticles Loaded with Polyphenolic Compounds of Moringa Oleifera Leaf Extract Have Hepatoprotective Activity in Bisphenol A-Induced Animal Models

Bisphenol A (BPA) is an xenoestrogenic chemical used extensively in the fabrication of baby bottles, reusable plastic water bottles and polycarbonate plastic containers. The current study aims to investigate the hepatoprotective activity of Moringa oleifera Lam leaf extract (MOLE) and hydrogel NPs made of starch-MOLE-Bovine Serum Albumin (BSA) against Bisphenol A-induced liver toxicity in male rats. Fabrication and characterization of hydrogel NPs formed of starch-MOLE-BSA were investigated using FTIR, TEM, zeta potential, UV-visible spectroscopy and fluorescence spectrophotometer. The potential efficacy of hydrogel NPs was studied. Compared to the results of control, the level of liver function, oxidative stress markers and lipid profile status were remodulated in the groups treated with MOLE and hydrogel NPs (Encap. MOLE). Meanwhile, the administration of MOLE and Encap MOLE significantly increased antioxidant activity and decreased the level of apoptotic pathways. Heme oxygenase (HO)-1 and growth arrest -DNA damage-inducible gene 45b (Gadd45b) were also regulated in the groups treated with MOLE and Encap. MOLE compared to the group which received BPA alone. In the present study, MOLE and hydrogel NPs led to remarkable alterations in histological changes during BPA administration. Overall, MOLE has a potential antioxidant activity which can be used in the treatment of liver disorders.

The Influence of Green and Black Tea Infusion Parameters on Total Polyphenol Content and Antioxidant Activity by ABTS and DPPH Assays

Tea contains about 230 chemical bioactive compounds, of which polyphenols represent the most considerable fraction (30% of total dry weight). These compounds have relevant nutritional and pharmacological effects on human health, exerting antioxidant activities against oxidative stress-induced damage. The industrial processes applied in tea production can lead to qualitative and quantitative changes in the phenolic content and composition and in antioxidant properties, thus influencing their potential biological activities. Meanwhile, the procedure for tea preparation may influence the quantity of the extracted phenolic compounds. In this study, the effects of different infusion parameters, such as the water type used for infusion (tap water, distilled water, and natural mineral water), time (3, 5, and 10 min), temperature (T = 80 °C and 100 °C), and pH (ranged between 3 and 9) were considered. The optimal infusion variables resulting from the study were obtained by extracting phenolic compounds at T = 100 °C for 10 min, both for green (916.12–1169.81 mg GAE/g) and black (932.03–1126.62 mg GAE/g) bagged tea samples, respectively.

Variable selection in the chemometric treatment of food data: A tutorial review

Food analysis covers aspects of quality and detection of possible frauds to ensure the integrity of the food. The arsenal of analytical instruments available for food analysis is broad and allows the generation of a large volume of information per sample. But this instrumental information may not yet give the desired answer; it must be processed to provide a final answer for decision making. The possibility of discarding non-informative and/or redundant signals can lead to models of better accuracy, robustness, and chemical interpretability, in line with the principle of parsimony. Thus, in this tutorial review, we cover aspects of variable selection in food analysis, including definitions, theoretical aspects of variable selection, and case studies showing the advantages of variable selection-based models concerning the use of a wide range of non-informative and redundant instrumental information in the analysis of food matrices.

Multielement authentication of apples from the cold highlands in southwest China

BACKGROUND

Half of all apple production worldwide comes from China. However, the geographic authentication of Chinese apples has not been well studied. We highlight the multi-element-based geographical discrimination of apples from the southwest cold highlands (SCH) of China. 565 samples from the SCH (138) and others (427) were obtained, and the content of fifteen elements were applied to construct models for discrimination.

RESULTS

The SCH apples from 2017 to 2019 had higher concentrations of Mn, Zn, Cr, Cd, Se, Pb, and Fe, but lower concentrations of Na, B, Ni, and P. With sufficient training, linear discriminant analysis (LDA) discriminated the SCH, and the testing accuracy averaged 92.5% and 92.2%. Nonlinear discrimination models were more suitable than the linear models. Optimized random forest analysis was the model with the best fit, and with averaged training and testing it obtained a level of accuracy of 98.2% and 98.5%.

CONCLUSION

The multielement-based discrimination of SCH apples could aid further studies of geographical origins. © 2021 Society of Chemical Industry.

Discrimination of Adulterated Ginkgo Biloba Products Based on 2T2D Correlation Spectroscopy in UV-Vis Range

Ginkgo biloba is a popular medicinal plant widely used in numerous herbal products, including food supplements. Due to its popularity and growing economic value, G. biloba leaf extract has become the target of economically motivated adulterations. There are many reports about the poor quality of ginkgo products and their adulteration, mainly by adding flavonols, flavonol glycosides, or extracts from other plants. In this work, we developed an approach using two-trace two-dimensional correlation spectroscopy (2T2D COS) in UV-Vis range combined with multilinear principal component analysis (MPCA) to detect potential adulteration of twenty G. biloba food supplements. UV-Vis spectral data are obtained for 80% methanol and aqueous extracts in the range of 245–410 nm. Three series of two-dimensional correlation spectra were interpreted by visual inspection and using MPCA. The proposed relatively quick and straightforward approach successfully differentiated supplements adulterated with rutin or those lacking ginkgo leaf extract. Supporting information about adulteration was obtained from the difference between the DPPH radical scavenging capacity of both extracts and from chromatographic (HPLC-DAD) fingerprints of methanolic samples.

Tea and Chicory Extract Characterization, Classification and Authentication by Non-Targeted HPLC-UV-FLD Fingerprinting and Chemometrics

Tea is a widely consumed drink in the world which is susceptible to undergoing adulterations to reduce manufacturing costs and rise financial benefits. The development of simple analytical methodologies to assess tea authenticity, as well as to detect and quantify frauds, is an important matter considering the rise of adulteration issues in recent years. In the present study, untargeted HPLC-UV and HPLC-FLD fingerprinting methods were employed to characterize, classify, and authenticate tea extracts belonging to different varieties (red, green, black, oolong, and white teas) by partial least squares-discriminant analysis (PLS-DA), as well as to detect and quantify adulteration frauds when chicory was used as the adulterant by partial least squares (PLS) regression, to ensure the authenticity and integrity of foodstuffs. Overall, PLS-DA showed a good classification and grouping of the tea samples according to the tea variety and, except for some white tea extracts, perfectly discriminated from the chicory ones. One hundred percent classification rates for the PLS-DA calibration models were achieved, except for green and oolong tea when HPLC-FLD fingerprints were employed, which showed classification rates of 96.43% and 95.45%, respectively. Good predictions were also accomplished, also showing, in almost all the cases, a 100% classification rate for prediction, with the exception of white tea and oolong tea when HPLC-UV fingerprints were employed that exhibited a classification rate of 77.78% and 88.89%, respectively. Good PLS results for chicory adulteration detection and quantitation were also accomplished, with calibration, cross-validation, and external validation errors beneath 1.4%, 6.4%, and 3.7%, respectively. Acceptable prediction errors (below 21.7%) were also observed, except for white tea extracts that showed higher errors which were attributed to the low sample variability available.

Development of a novel green tea quality roadmap and the complex sensory-associated characteristics exploration using rapid near-infrared spectroscopy technology

Nondestructive instrumental identification of the green tea quality instead of professional human panel tests is highly desired for industrial application recently. The special flavor is a key quality-trait that influence consumer preference. However, flavonoids, as well as sensory-associated compounds, which play a critical role in the quality-traits profile of green tea samples have been poorly investigated. In this study, we were proposing an objective and accurate near infrared spectroscopy (NIRS) profile to support quality control within the entire green tea sensory evaluation chain, the complexity of green tea samples' sensory analysis was performed by two complementary methods: the standard calculation and the novel NIRS roadmap coupled with chemometrics. The green tea samples' physical quality, gustatory index, and nutritional index were measured respectively, which taking into consideration the gustatory evaluation of green tea for five commercially representative overall quality ("very bad", "bad", "regular", "good" and "excellent"). Our findings highlight the underexplored role of NIRS in chemical-to-sensory relationships and its widespread importance and utility in green tea quality improvement. Collectively, the comprehensive characterization of sensory-associated attribution allowed the identification of a wide array of spectrometric features, mostly related to moisture, soluble solids (SS), tea polyphenol (TPP), epigallocatechin gallate (EGCG), epicatechin (EC) and tea polysaccharide (TPS), which can be used as putative biomarkers to rapidly evaluate the green tea flavor variations related to rank differences. Otherwise, the NIRS' data were split into the calibration (n = 80) and prediction (n = 40) set independently, which showed high correlation coefficient with Rp-values of 0.9024, 0.9020 in physical and total cup quality, respectively. In this research, we demonstrated that NIRS was an easily-generated strategy and able to close the loop to feedback into the process for advanced process control. However, the established models should be improved by more green tea samples from different regions.

Honey authentication in terms of its adulteration with sugar syrups using UV-Vis spectroscopy and one-class classifiers

This work proposes the use of UV-Vis spectroscopy and one-class classifiers to authenticate honey in terms of their individual and simultaneous adulterations with corn syrup, agave syrup, and sugarcane molasses. Then, spectra of aqueous authentic (n = 73) and adulterated (n = 162) honey samples were recorded. Before the construction of OC-PLS and DD-SIMCA models, different pre-processing techniques were used to removed baseline shifts. The best result obtained by DD-SIMCA using offset correction, correctly classifying all the samples in the test set. Therefore, the proposed methodology can be used as a promising tool to authenticate honey and prevent fraudulent labeling, affording security to consumers and providing an alternative to regulatory agencies. Moreover, it avoids laborious sample preparation and additional operational costs, since the analytical information is acquired using a routine instrumental technique, without the need for any sample preparation step, other than dilution of the samples in water alone.

Scores selection via Fisher's discriminant power in PCA-LDA to improve the classification of food data

This paper proposes an adaptation of the Fisher's discriminability criterion (named here as discriminant power, DP) for choosing principal components (obtained from Principal Component Analysis, PCA), which will be used to construct supervised Linear Discriminant Analysis (LDA) models for solving classification problems of food data. The proposed PCA-DP-LDA algorithm was then applied to (i) simulated data, (ii) classify soybean oils with respect to expiration date, and (iii) identify cachaça adulteration with wood extracts that simulated aging. For comparison, PCA-DP-LDA was evaluated against conventional PCA-LDA (based on explained variance) and Partial Least Squares-Discriminant Analysis (PLS-DA). Among them, PCA-DP-LDA achieved the most parsimonious and interpretable results, with similar or better classification performance. Therefore, the new algorithm can be considered a good alternative to the already well-established discriminant methods, being potentially applied where the discriminability of the principal components may not follow the same behavior of the explained variance.

Development of ANN models based on combined UV-vis-NIR spectra for rapid quantification of physical and chemical properties of industrial hemp extracts

OBJECTIVES

The aim of this study was to develop artificial neural network (ANNs) models for prediction of physical (total dissolved solids, extraction yield) and chemical (total polyphenolic content, antioxidant activity) properties of industrial hemp extracts, prepared by two different extraction methods (solid-liquid extraction and microwave-assisted extraction) based on combined UV-VIS-NIR spectra. Spectral data were gathered for 46 samples per extraction method.

RESULTS

The PCA analysis ensured efficient separation of the samples based on the amount of ethanol in extraction solvent using NIR spectra for both conventional and microwave-assisted extraction.

CONCLUSIONS

Results showed that reliable ANN models (R² >0.7000) for describing physical, chemical, and simultaneously physical and chemical characteristics can be developed based on combined UV-VIS-NIR spectra of industrial hemp extracts without spectra pre-processing.

Emerging techniques for determining the quality and safety of tea products: A review

Spectroscopic techniques, electrochemical methods, nanozymes, computer vision, and modified chromatographic techniques are the emerging techniques for determining the quality and safety parameters (e.g., physical, chemical, microbiological, and classified parameters, as well as inorganic and organic contaminants) of tea products (such as fresh tea leaves, commercial tea, tea beverage, tea powder, and tea bakery products) effectively. By simplifying the sample preparation, speeding up the detection process, reducing the interference of other substances contained in the sample, and improving the sensitivity and accuracy of the current standard techniques, the abovementioned emerging techniques achieve rapid, cost-effective, and nondestructive or slightly destructive determination of tea products, with some of them providing real-time detection results. Applying these emerging techniques in the whole industry of tea product processing, right from the picking of fresh tea leaves, fermentation of tea leaves, to the sensory evaluation of commercial tea, as well as developing portable devices for real-time and on-site determination of classified and safety parameters (e.g., the geographical origin, grade, and content of contaminants) will not only eliminate the strong dependence on professionals but also help mechanize the production of tea products, which deserves further research. Conducting a review on the application of spectroscopic techniques, electrochemical methods, nanozymes, computer vision, and modifications of chromatographic techniques for quality and safety determination of tea products may serve as guide for other types of foods and beverages, offering potential techniques for their detection and evaluation, which would promote the development of the food industry.

Robust Classification of Tea Based on Multi-Channel LED-Induced Fluorescence and a Convolutional Neural Network

A multi-channel light emitting diode (LED)-induced fluorescence system combined with a convolutional neural network (CNN) analytical method was proposed to classify the varieties of tea leaves. The fluorescence system was developed employing seven LEDs with spectra ranging from ultra-violet (UV) to blue as excitation light sources. The LEDs were lit up sequentially to induce a respective fluorescence spectrum, and their ability to excite fluorescence from components in tea leaves were investigated. All the spectral data were merged together to form a two-dimensional matrix and processed by a CNN model, which is famous for its strong ability in pattern recognition. Principal component analysis combined with k-nearest-neighbor classification was also employed as a baseline for comparison. Six grades of green tea, two types of black tea and one kind of white tea were verified. The result proved a significant improvement in accuracy and showed that the proposed system and methodology provides a fast, compact and robust approach for tea classification.

HPTLC-PCA Complementary to HRMS-PCA in the Case Study of Arbutus unedo Antioxidant Phenolic Profiling

A comparison between High-Performance Thin-Layer Chromatography (HPTLC) analysis and Liquid Chromatography High Resolution Mass Spectrometry (LC-HRMS), coupled with Principal Component Analysis (PCA) was carried out by performing a combined metabolomics study to discriminate Arbutus unedo (A. unedo) plants. For a rapid digital record of A. unedo extracts (leaves, yellow fruit, and red fruit collected in La Maddalena and Sassari, Sardinia), HPTLC was used. Data were then analysed by PCA with the results of the ability of this technique to discriminate samples. Similarly, extracts were acquired by non-targeted LC-HRMS followed by unsupervised PCA, and then by LC-HRMS (MS) to identify secondary metabolites involved in the differentiation of the samples. As a result, we demonstrated that HPTLC may be applied as a simple and reliable untargeted approach to rapidly discriminate extracts based on tissues and/or geographical origins, while LC-HRMS could be used to identify which metabolites are able to discriminate samples.

Alternative data mining/machine learning methods for the analytical evaluation of food quality and authenticity - A review

In recent years, the variety and volume of data acquired by modern analytical instruments in order to conduct a better authentication of food has dramatically increased. Several pattern recognition tools have been developed to deal with the large volume and complexity of available trial data. The most widely used methods are principal component analysis (PCA), partial least squares-discriminant analysis (PLS-DA), soft independent modelling by class analogy (SIMCA), k-nearest neighbours (kNN), parallel factor analysis (PARAFAC), and multivariate curve resolution-alternating least squares (MCR-ALS). Nevertheless, there are alternative data treatment methods, such as support vector machine (SVM), classification and regression tree (CART) and random forest (RF), that show a great potential and more advantages compared to conventional ones. In this paper, we explain the background of these methods and review and discuss the reported studies in which these three methods have been applied in the area of food quality and authenticity. In addition, we clarify the technical terminology used in this particular area of research.

Tea types classification with data fusion of UV-Vis, synchronous fluorescence and NIR spectroscopies and chemometric analysis

The potential of selected spectroscopic methods - UV-Vis, synchronous fluorescence and NIR as well a data fusion of the measurements by these methods - for the classification of tea samples with respect to the production process was examined. Four classification methods - Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Regularized Discriminant Analysis (RDA) and Support Vector Machine (SVM) - were used to analyze spectroscopic data. PCA analysis was applied prior to classification methods to reduce multidimensionality of the data. Classification error rates were used to evaluate the performance of these methods in the classification of tea samples. The results indicate that black, green, white, yellow, dark, and oolong teas, which are produced by different methods, are characterized by different UV-Vis, fluorescence, and NIR spectra. The lowest error rates in the calibration and validation data sets for individual spectroscopies and data fusion models were obtained with the use of the QDA and SVM methods, and did not exceed 3.3% and 0.0%, respectively. The lowest classification error rates in the validation data sets for individual spectroscopies were obtained with the use of RDA (12,8%), SVM (6,7%), and QDA (2,7%), for the UV-Vis, SF, and NIR spectroscopies, respectively. NIR spectroscopy combined with QDA outperformed other individual spectroscopic methods. Very low classification errors in the validation data sets - below 3% - were obtained for all the data fusion data sets (SF + UV-Vis, SF + NIR, NIR + UV-Vis combined with the SVM method). The results show that UV-Vis, fluorescence and near infrared spectroscopies may complement each other, giving lower errors for the classification of tea types.

Combining excitation-emission matrix fluorescence spectroscopy, parallel factor analysis, cyclodextrin-modified micellar electrokinetic chromatography and partial least squares class-modelling for green tea characterization

In this study, an alternative analytical approach for analyzing and characterizing green tea (GT) samples is proposed, based on the combination of excitation-emission matrix (EEM) fluorescence spectroscopy and multivariate chemometric techniques. The three-dimensional spectra of 63 GT samples were recorded using a Perkin-Elmer LS55 luminescence spectrometer; emission spectra were recorded between 295 and 800 nm at excitation wavelength ranging from 200 to 290 nm, with excitation and emission slits both set at 10 nm. The excitation and emission profiles of two factors were obtained using Parallel Factor Analysis (PARAFAC) as a 3-way decomposition method. In this way, for the first time, the spectra of two main fluorophores in green teas have been found. Moreover, a cyclodextrin-modified micellar electrokinetic chromatography method was employed to quantify the most represented catechins and methylxanthines in a subset of 24 GT samples in order to obtain complementary information on the geographical origin of tea. The discrimination ability between the two types of tea has been shown by a Partial Least Squares Class-Modelling performed on the electrokinetic chromatography data, being the sensitivity and specificity of the class model built for the Japanese GT samples 98.70% and 98.68%, respectively. This comprehensive work demonstrates the capability of the combination of EEM fluorescence spectroscopy and PARAFAC model for characterizing, differentiating and analyzing GT samples.

"Turn-off" fluorescent sensor based on double quantum dots coupled with chemometrics for highly sensitive and specific recognition of 53 famous green teas

Fluorescent "turn-off" sensors based on double quantum dots (QDs) has attracted increasing attention in the detection of many materials due to their properties such as more useful information, higher fluorescence efficiency and stability compared with the fluorescent "turn-off" sensors based on single QDs. In this work, highly sensitive and specific method for recognition of 53 different famous green teas was developed based on the fluorescent "turn-off" model with water-soluble ZnCdSe-CdTe double QDs. The fluorescence of the two QDs can be quenched by different teas with varying degrees, which results in the differences in positions and intensities of two peaks. By the combination of classic partial least square discriminant analysis (PLSDA), all the green teas can be discriminated with high sensitivity, specificity and a satisfactory recognition rate of 100% for training set and 100% for prediction set, respectively. The fluorescent "turn-off" sensors based on the single QDs (either ZnCdSe QDs or CdTe QDs) coupled with PLSDA were also employed to recognize the 53 famous green teas with unsatisfactory results. Therefore, the fluorescent "turn-off" sensors based on the double QDs is more appropriate for the large-class-number classification (LCNC) of green teas. Herein, we have demonstrated, for the first time, that so many kinds of famous green teas can be discriminated by the "turn-off" model of double QDs combined with chemometrics, which has largely extended the capability of traditional fluorescence and chemometrics, as well as exhibits great potential to perform LCNC in other practical applications.

Partial least squares-discriminant analysis (PLS-DA) for classification of high-dimensional (HD) data: a review of contemporary practice strategies and knowledge gaps

This review highlights and discusses critically various knowledge gaps in classification modelling using PLS-DA for high dimensional data.

Characterisation of electrospun gelatine nanofibres encapsulated with Moringa oleifera bioactive extract

BACKGROUND

Drumstick (Moringa oleifera) leaves have been used as a folk herbal medicine across many cultures since ancient times. This is most probably due to presence of phytochemicals possessing antioxidant properties, which could retard oxidative stress, and their degenerative effect. The current study deals with nanoencapsulation of Moringa oleifera (MO) leaf ethanolic extract within fish sourced gelatine matrix using electrospinning technique.

RESULTS

The total phenolic and flavonoid content, radical scavenging (IC₅₀ ) and metal reducing properties were 67.0 ± 2.5 mg GAE g^-1 sample 32.0 ± 0.5 mg QE g^-1 extract, 0.08 ± 0.01 mg mL^-1 and 510 ± 10 µmol eq Fe(II) g^-1 extract, respectively. Morphological and spectroscopic analysis of the fibre mats confirmed successful nanoencapsulation of MO extract within defect free nanofibres via electrospinning process. The percentage encapsulation efficiency (EE) was between 80% and 85%. Furthermore, thermal stability of encapsulated fibres, especially at 3% and 5% of core loading content, was significantly improved. Toxicological analysis revealed that the extract in its original and encapsulated form was safe for oral consumption.

CONCLUSION

Overall, the present study showed the potential of ambient temperature electrospinning process as a safe nanoencapsulation method, where MO extract retained its antioxidative capacities. © 2016 Society of Chemical Industry.

Experimental variability and data pre-processing as factors affecting the discrimination power of some chemometric approaches (PCA, CA and a new algorithm based on linear regression) applied to (+/-)ESI/MS and RPLC/UV data: Application on green tea extracts

The influence of the experimental variability (instrumental repeatability, instrumental intermediate precision and sample preparation variability) and data pre-processing (normalization, peak alignment, background subtraction) on the discrimination power of multivariate data analysis methods (Principal Component Analysis -PCA- and Cluster Analysis -CA-) as well as a new algorithm based on linear regression was studied. Data used in the study were obtained through positive or negative ion monitoring electrospray mass spectrometry (+/-ESI/MS) and reversed phase liquid chromatography/UV spectrometric detection (RPLC/UV) applied to green tea extracts. Extractions in ethanol and heated water infusion were used as sample preparation procedures. The multivariate methods were directly applied to mass spectra and chromatograms, involving strictly a holistic comparison of shapes, without assignment of any structural identity to compounds. An alternative data interpretation based on linear regression analysis mutually applied to data series is also discussed. Slopes, intercepts and correlation coefficients produced by the linear regression analysis applied on pairs of very large experimental data series successfully retain information resulting from high frequency instrumental acquisition rates, obviously better defining the profiles being compared. Consequently, each type of sample or comparison between samples produces in the Cartesian space an ellipsoidal volume defined by the normal variation intervals of the slope, intercept and correlation coefficient. Distances between volumes graphically illustrates (dis)similarities between compared data. The instrumental intermediate precision had the major effect on the discrimination power of the multivariate data analysis methods. Mass spectra produced through ionization from liquid state in atmospheric pressure conditions of bulk complex mixtures resulting from extracted materials of natural origins provided an excellent data basis for multivariate analysis methods, equivalent to data resulting from chromatographic separations. The alternative evaluation of very large data series based on linear regression analysis produced information equivalent to results obtained through application of PCA an CA.