Classification of specialty coffees using machine learning techniques (original) (raw)
Related papers
2020
Brazil is the largest exporter of coffee beans, 29% world exports, 15% this volume in specialty coffees. Thereby researches are done, so that identify different segments in the market, in order to direct the end consumer to a better quality product. New technologies are explored to meet an increasing demand for high quality coffees. Therefore, in this article has an objective to propose the use of machine learning techniques combined with projection pursuit in the construction of unsupervised classification models, in a sensory acceptance experiment, applied to four groups of trained and untrained consumers, in four classes of specialty coffees in which they were evaluated sensory characteristics: aroma, body coffee, sweetness and general note. For evaluating classifier performance, in the data with reduced dimension, all instances were used, and considering four groupings, the models were adjusted. The results obtained from the groupings formed were compared with pre-established cl...
Recent Advances on Numerical Simulations [Working Title], 2020
The sensory analysis of coffees assumes that a sensory panel is formed by tasters trained according to the recommendations of the American Specialty Coffee Association. However, the choice that routinely determines the preference of a coffee is made through experimentation with consumers, in which, for the most part, they have no specific ability in relation to sensory characteristics. Considering that untrained consumers or those with basic knowledge regarding the quality of specialty coffees have little ability to discriminate between different sensory attributes, it is reasonable to admit the highest score given by a taster. Given this fact, probabilistic studies considering appropriate probability distributions are necessary. To access the uncertainty inherent in the notes given by the tasters, resampling methods such as Monte Carlo’s can be considered and when there is no knowledge about the distribution of a given statistic, p-Bootstrap confidence intervals become a viable alt...
Sensory quality prediction of coffee assessed by physicochemical parameters and Multivariate model
Coffee Science, 2020
Beverages from roasted coffee can be classified according to their sensory quality into Gourmet, Superior, Traditional, and not recommended for supply coffees. However, the sensory evaluation of coffee has been questioned as it can induce a subjective bias, since the assessors may be influenced by psychological, physiological, and/or emotional factors. Therefore, the aim of this study was to develop multivariate models for predicting the overall quality of Gourmet, Superior, and Traditional coffees, based on the physical and physicochemical parameters. One hundred and eight ground roasted coffee samples were evaluated for particle size, degree of roasting, histological identification, moisture, ash, aqueous extract, soluble solids (Brix), pH, and sensory profiling. All categories presented fine grinding. No significant differences were observed in the moisture content and soluble solids (Brix) of Gourmet, Superior, Traditional, not recommended for supply coffee samples. The Traditional and not recommended for supply presented higher levels of aqueous extract, ash, and pH. Light degree of roast and higher acidity values were observed with the increase in coffee quality grades. The results of the physical and physicochemical parameters and the principal component analysis allowed the separation of coffees into only two classes: high-quality (Gourmet and Superior) and low-quality (Traditional and not recommended). Furthermore, the one-class classification (OCC) method showed good sensitivity and was able to satisfactorily distinguish the Gourmet coffee samples from the other samples, in this way, this model can be used to corroborate but not replace the sensory analysis.
Sustainability
This study investigates the classification of Arabic coffee into three major variations (light, medium, and dark) using simulated data gathered from the actual measurements of color information, antioxidant laboratory testing, and chemical composition tests. The goal is to overcome the restrictions of limited real-world data availability and the high costs involved with laboratory testing. The Monte Carlo approach is used to generate new samples for each type of Arabic coffee using the mean values and standard deviations of publicly available data. Using these simulated data, multiple machine-learning algorithms are used to classify Arabic coffee, while also investigating the importance of features in identifying the key chemical components. The findings emphasize the importance of color information in accurately recognizing Arabic coffee types. However, depending purely on antioxidant information results in poor classification accuracy due to increased data complexity and classifie...
Food Science and Technology, 2021
Coffee is one of the most popular beverages in the world, and changes in production, processing, trading, appreciation, and the culture of consumers are noticeable (Guimarães et al., 2019). Specialty and high-quality coffees are gaining space in the market, meeting the demands of consumers (Giacalone et al., 2019; Ufer et al., 2019). The worldwide consumption of specialty coffee is growing (with an increase of 1.5%), and this growth in the market is affected by new products, research, and specialized coffee shops (Guimarães et al., 2019). For these reasons, it is necessary to evaluate consumers' behavior and desires (Wang & Yu, 2016), detecting a large number of markets that can be explored with greater accuracy and quality. Market surveys are essential for understanding consumer intentions. The use of the internet (twitter, e-mail, among others) in the application of questionnaires using different qualitative sensory methods (such as completion task) has become very important to discover the motivations, perceptions, and attitudes of consumers about a product (Sass et al., 2020; Torres et al., 2020). However, there is not enough information about the specialty coffee market, especially about the characteristics and buying behavior of consumers (Guimarães et al., 2019). Several studies have shown that extrinsic factors, such as packaging, brand, information, emotion, and atmospheric, influence the sensory perception of food products (Spence, 2015;
Food Quality and Preference, 2002
Grading systems since their conception have aimed at facilitating commercialisation of food stuff world-wide. However, while many food products can have their quality assessed by analytical means, there are many foods that are sold according to their sensory quality and for which quality is not easily measured by conventional analytical techniques. Measuring sensory quality in some products has moved forward and utilises fully trained panels to set-up quality control systems and routine evaluations while others still rely on traditional commercial classifications and grading systems. The grading discussed here must be differentiated from grading using ''trained experts'' to evaluate food products according to legislated standards. There are specific cases in which the need to comply with national or international standards requires the development of specially trained tasting experts. Bisogni, [In D. E. Kramer, & J. Liston (Eds.), Seafood quality determination (p. 547). Amsterdam: Elsevier Science] illustrates quite well this specific scenario. The case of coffee, the most traded agricultural crop world-wide is examined here. In coffee a high diversity of classification systems is applied and the use of the ''expert cupper'' is the norm. There is not a unique and universal system applied world-wide for the quality control of green coffee. Tailor made procedures are selectively implemented by International, National, local bodies, trading institutions and private companies. Procedures are mostly geared to facilitate the trading of the commodity and sensory quality is in most cases described by ''cuppers'' or ''liquorers'' using personal opinion and tasting experience accumulated over the years. #
Discrimination of Civet Coffee Using Image Processing and Machine Learning
ABSTRACT This paper is about separating the two classes of civet coffee with the non-civet coffee. Image processing was used to extract the color and texture features of the civet and non-civet coffee, namely the red, green, blue, hue , saturation, brightness, entropy, energy, contrast, homogeneity. The 23 classifiers of MATLAB Classification Learner APP were used to classify the ten features. Among the 23 classifiers, the best in terms of accuracy is the quadratic support vector machine which achieved 79.7 % accuracy.
Foods
The Specialty Coffee Association (SCA) sensory analysis protocol is the methodology that is used to classify specialty coffees. However, because the sensory analysis is sensitive to the taster’s training, cognitive psychology, and physiology, among other parameters, the feasibility of instrumental approaches has been recently studied for complementing such analyses. Spectroscopic methods, mainly near infrared (NIR) and mid infrared (FTIR—Fourier Transform Infrared), have been extensively employed for food quality authentication. In view of the aforementioned, we compared NIR and FTIR to distinguish different qualities and sensory characteristics of specialty coffee samples in the present study. Twenty-eight green coffee beans samples were roasted (in duplicate), with roasting conditions following the SCA protocol for sensory analysis. FTIR and NIR were used to analyze the ground and roasted coffee samples, and the data then submitted to statistical analysis to build up PLS models in...
Acta Scientiarum. Agronomy
The identification and interpretation of discrepant observations in sensory experiments are difficult to implement since the external effects are associated with the individual consumer. This fact becomes more relevant in experiments that involve blends, which scrutinize coffees with different qualities, varieties, origins, and forms of processing and preparation. This work proposes a statistical procedure that facilitates the identification of outliers while also evaluating the discriminatory powers of a sensory panel concerning the differentiation of pure blends and coffees. For this purpose, four experiments were performed that tested coffees with different qualities and varieties. The results suggest that the statistical procedure proposed in this work was effective for discriminating the blends relative to the pure coffees and that the effects of the concentrations and types of processing did not interfere with the statistical evaluations.
Tasters’ performance in a coffee quality contest in Brazil
Coffee Science, 2021
The objective of this work was to evaluate the performance of coffee tasters in five annual editions of Minas Gerais Coffee Quality Contest. The repeatability coefficients of the tasters' scores for sensory attributes were estimated, as well as the minimum numbers of tasters required for consistent sensory results, and the groups of tasters by (dis)similarity of sensory scores. For the repeatability analysis, the treatments (coffees) were tested with the repetitions, constituted by the tasters. The repeatability coefficients were estimated using the analysis of variance, principal component and structural analysis methods. The minimum number of tasters was obtained based on pre-established determination coefficients. Euclidean distance matrices between tasters were determined, which were used as a measure of dissimilarity for cluster analysis by the Tocher optimization method. The tasters' performance in five annual editions of Minas Gerais Coffee Quality Contest is reliable using COE or SCA sensory analysis protocols. Although not fully calibrated, most tasters are grouped with similar cupping results. Unless efficient calibration prior to the contest is adopted, the number of tasters to be used in the next contest editions can not be drastically and randomly reduced, since the estimated minimum number varied over the years. Calibration activities are suggested to improve two main aspects of the Minas Gerais Coffee Quality Contest: distinguishing the best coffees and trainning tasters.