Predict Wine Quality Research Articles

Performance Evaluation of Multiple Deep Learning Models for Wine Quality Prediction

Research utilizing a dataset from the UCI repository evaluated the predictive accuracy of nine machine learning models for wine quality. The models employed include Logistic Regression, K-Nearest Neighbor (KNN), Decision Tree, Support Vector Machine (SVM), Random Forest, XGBoost, LightGBM, CatBoost, and Gradient Boosting. The dataset comprises 1,599 samples with 12 chemical parameters. Data preprocessing, including oversampling, normalization, standardization, and seeding, was performed to enhance model performance.The study's findings indicate that the models with the highest accuracy values were LightGBM (87.80%), CatBoost (86.60%), and Random Forest (85.70%). A voting classifier combining these three models achieved an accuracy of 87.29%. Further analysis using a confusion matrix demonstrated that this combined model effectively predicts the "Good" and "Not Good" classes.In conclusion, the combination of LightGBM, CatBoost, and Random Forest models proves to be an effective approach for predicting wine quality based on chemical parameters, with an accuracy value of 87.29%.

Towards Predicting the Quality of Red Wine Using Novel Machine Learning Methods for Classification, Data Visualization, and Analysis

There is a growing concern among consumers and the wine industry regarding the quality of wine. Traditionally, wine experts determined its quality through tasting, which was time-consuming. Therefore, there is a need to predict wine quality based on specific key features to streamline these tasks. Technological developments like machine learning approaches have replaced human assessments with computational methods. However, some of these methods have faced criticism due to their low accuracy and lack of interpretability for humans. In this paper, a stacking ensemble method is introduced and demonstrates superior predictive performance when compared to other classification techniques like logistic regression, decision trees, gradient boosting, adaptive boosting (AdaBoost), and random forest. This evaluation is based on classification metrics such as accuracy, precision, recall, and F1-Score, all under the same conditions. Additionally, outlier detection algorithms were employed to identify exceptional or subpar wines, though their results did not match the accuracy of classification approaches. Lastly, a feature analysis study was conducted to assess the significance of each feature in the model’s performance. Received: 3 November 2023 | Revised: 21 February 2024 | Accepted: 28 April 2024 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement The data that support the findings of this study are openly available in UCI machine learning repository at https://archive.ics.uci.edu/ml/datasets/wine+quality. The technical details, code, and data can be accessed by visiting the link: https://github.com/jovialniyo93/red_wine_quality_prediction. Author Contribution Statement Jovial Niyogisubizo: Conceptualization, Methodology, Software, Validation, Formal analysis, Data curation, Writing - original draft, Writing - review & editing, Project administration. Jean de Dieu Ninteretse: Validation, Formal analysis, Writing - original draft, Writing - review & editing. Eric Nziyumva: Validation, Formal analysis, Resources, Writing - original draft, Writing - review & editing, Supervision. Marc Nshimiyimana: Validation, Formal analysis, Investigation, Writing - original draft, Writing - review & editing, Visualization. Evariste Murwanashyaka: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing - original draft, Writing - review & editing, Visualization, Supervision. Erneste Habiyakare: Validation, Formal analysis, Investigation, Resources, Writing - original draft, Writing - review & editing, Visualization, Supervision.

RED WINE QUALITY PREDICTION USING MACHINE LEARNING

Wine quality assessment is a critical task in the beverage industry, as it directly impacts consumer satisfaction and market competitiveness. Among various types of wine, red wine stands out for its complexity in flavor, aroma, and texture, making its quality prediction a challenging yet crucial endeavor. This project aims to develop a machine learning model for predicting the quality of red wine based on its physicochemical properties. The dataset utilized for this project consists of various attributes such as acidity, pH, alcohol content, and volatile acidity, among others, collected from red wine samples. These attributes serve as input features for the machine learning model, while the quality of wine, typically rated on a scale, serves as the target variable. The proposed approach involves several stages: data preprocessing, feature selection, model selection, and evaluation. During data preprocessing, techniques like normalization and handling missing values are employed to ensure data quality. Feature selection techniques such as correlation analysis and feature importance are utilized to identify the most relevant attributes for predicting wine quality. Several machine learning algorithms, including but not limited to, linear regression, decision trees, random forests, and support vector machines, are trained and evaluated to determine the most suitable model for the task. Performance metrics such as mean squared error, accuracy, and F1 score are utilized to assess the model's predictive capability. Key Words: Wine quality, market, aroma, texture, physicochemical, alcohol, acidity.

Using Machine Learning to Predict Wine Quality and Prices: A Demonstrative Case Using a Large Tabular Database

Using Machine Learning to Predict Wine Quality and Prices: A Demonstrative Case Using a Large Tabular Database

Development and Comparison of Regression Models Predicting Wine Quality using Excel - From a Profit Maximization Perspective

Wine reviewers and the general public usually determine the quality of the wine. There is a great deal of subjec-tivity in such reviews and differences in the reviewers' preferences, making it difficult for the winery to obtain favorable information. This paper will use Excel to build a related model of chemical substances in wine and wine quality. Both linear and logistic regression is used in this article to predict wine quality. In addition, differ-ent from past literature that indicates the quality of red wine based on chemical substances, this paper creatively constructs a profit model based on predicting wine quality. It thus helps wine sellers to make model selections. All these can help producers understand how to make good wine and get higher profits.

Analyzing of salient features and classification of wine type based on quality through various neural network and support vector machine classifiers

Wine quality certification is crucial to the wine industry. Indian wine’s superior quality is well-known around the world. Wine quality certification is crucial to the wine industry. Our main objective in this study is to find out a machine-learning model based on experimental data that has been gathered from various places across India and available synthetic data in order to predict wine quality. We utilized 178 wine samples with 13 different physiochemical characteristics. All the features have been analyzed and shown the values of max, min, mean, Kurt, skewness, and standard deviation of each variable. Important attributes that can be selected by comparing the values of all the above-mentioned feature selection approaches were required very much to improve wine quality. Five neural network methods and six support vector methods were trained and tested on all features of the dataset. Narrow neural network, Wide neural network, Quadratic support vector machine and Medium Gaussian support vector machine — all these classifiers showed 97.8% accuracy when trained and evaluated with all features but Quadratic support vector machine achieved this accuracy with the lowest training time 0.92556 sec ad the highest prediction speed 6400 obs/sec. These results demonstrate that the anticipated and experimental responses are extremely well aligned, according to this accuracy and the preferred model is appropriate to classify wine quality based on physiochemical components.

Using ROC-curve to Illustrate the Use of Binomial Logistic Regression Model in Wine Quality Analysis

The major objective of this paper is to predict wine quality based on chemical attributes using binary logistic regression model. The evaluation of wine quality in this study is analyzed based on statistical methods, correlation analysis, and regression analysis. This study highlights several features are not relevant (only six essential variables were enough) to predict wine quality. The result showed that this method is robust enough to predict wine quality with an overall accuracy of more than 75%. Similar techniques or machine learning methods can help in predicting more accurately.

Understanding Quality of Pinot Noir Wine: Can Modelling and Machine Learning Pave the Way?

Wine research has as its core components the disciplines of sensory analysis, viticulture, and oenology. Wine quality is an important concept for each of these disciplines, as well as for both wine producers and consumers. Any technique that could help producers to understand the nature of wine quality and how consumers perceive it, will help them to design even more effective marketing strategies. However, predicting a wine’s quality presents wine science modelling with a real challenge. We used sample data from Pinot noir wines from different regions of New Zealand to develop a mathematical model that can predict wine quality, and applied dimensional analysis with the Buckingham Pi theorem to determine the mathematical relationship among different chemical and physiochemical compounds. This mathematical model used perceived wine quality indices investigated by wine experts and industry professionals. Afterwards, machine learning algorithms are applied to validate the relevant sensory and chemical concepts. Judgments of wine intrinsic attributes, including overall quality, were made by wine professionals to two sets of 18 Pinot noir wines from New Zealand. This study develops a conceptual and mathematical framework to predict wine quality, and then validated these using a large dataset with machine learning approaches. It is worth noting that the predicted wine quality indices are in good agreement with the wine experts’ perceived quality ratings.

Analysis of Machine Learning Algorithm to predict Wine Quality.

Product quality certification is used by industries to sell or advertise their products. The quality of wine is assessed by a human specialist, which is a time-consuming process that makes it quite expensive. Several machine learning techniques have already been applied to evaluate wine qualities such as quality and class on wine quality datasets. The quality of wine is determined not only by the amount of alcohol in it, but also by many traits, which change through time and therefore refine the wine's quality. It is critical to establish the wine's quality and categorise it into several categories based on a quality assessment. This study employs a variety of machine learning algorithms to predict wine quality. This research gives a comparison of fundamental and technical analysis based on many characteristics. This research compares and contrasts several prediction algorithms used to predict wine quality. Technical analysis such as time series analysis and machine learning algorithms like Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, Ada Boost Classifier, and Gradient Boosting Classifier are examples of these methodologies. With the use of visualisation, several techniques are evaluated based on methodologies, datasets, and efficiency.

A machine learning application in wine quality prediction

The wine business relies heavily on wine quality certification. The excellence of New Zealand Pinot noir wines is well-known worldwide. Our major goal in this research is to predict wine quality by generating synthetic data and construct a machine learning model based on this synthetic data and available experimental data collected from different and diverse regions across New Zealand. We utilised 18 Pinot noir wine samples with 54 different characteristics (7 physiochemical and 47 chemical features). We generated 1381 samples from 12 original samples using the SMOTE method, and six samples were preserved for model testing. The findings were compared using four distinct feature selection approaches. Important attributes (referred as essential variables) that were shown to be relevant in at least three feature selection methods were utilised to predict wine quality. Seven machine learning algorithms were trained and tested on a holdout original sample. Adaptive Boosting (AdaBoost) classifier showed 100% accuracy when trained and evaluated without feature selection, with feature selection (XGB), and with essential variables (features found important in at least three feature selection methods). In the presence of essential variables, the Random Forest (RF) classifier performance was increased.

Smart Agri Wine: An Artificial Intelligence Approach to Predict Wine Quality

Lately the item quality has been one of the critical parts of each and every industry. The customary strategies for surveying the item quality are exceptionally tedious and furthermore not having the ideal outcome with the resultant in unique Technology development. Through the ideas of Artificial Intelligence (AI) and Information Science (IS) it is more productive to evaluate or to foresee any sort of thing effectively. In this study, the investigation of wine information is done on University of California Irvine (UCI) Machine Learning (ML) dataset. The fundamental motivation behind this examination is to foresee wine quality dependent on physicochemical information using AI.

A Study and Analysis of Machine Learning Techniques in Predicting Wine Quality

In today’s trend consumers are very much concern about the quality of the product in turn, Industries are all working on various methodologies to ensure the high quality in their products. Most of consumers judge the quality of the product based on the certification obtained for the product. In Earlier days, the quality is measured and validated only through human experts. Nowadays most of the validation tasks are automated through software and this ease the burden of human experts by assisting with them in predicting the quality of the product and that leads to greater a reduction of time spent. Wine consumption has increased rapidly over the last few decades, not only for recreational purposes but also due of its inherent health benefits especially to human heart. This chapter demonstrates the usage of various machine learning techniques in predicting the quality of wine and results are validated through various quantitative metrics. Moreover the contribution of various independent variables facilitating the final outcome is precisely portrayed.

A machine learning approach using random forest and LASSO to predict wine quality

Quality assessment is a key factor for the wine industry, where the aim is to meet consumers' needs/demands and promote sales. Quality assessment is usually performed by experts and it is a time-consuming and expensive process. This paper proposes an alternative assessment using machine learning methods, such as the least absolute shrinkage and selection operator (LASSO) and random forest to predict wine quality. Our data analysis is based on a real wine dataset provided by a well-known wine firm in Greece. For this purpose, we employ the LASSO method, which is particularly effective in selecting the best possible number of variables required. Additionally, the random forest method is used and its findings are contrasted to those derived by four different M.L. methods, namely, linear discriminant analysis (LDA), classification and regression trees (CART), k-nearest neighbours (kNN) and support vector machines (SVM), and using the well-known ten-fold cross-validation method. The results of our analysis show that the statistical technique of random forest proposed improves the accuracy of the prediction wine quality, up to almost 95%, compared to the rankings attributed by wine tasters.

A machine learning approach using random forest and LASSO to predict wine quality

A machine learning approach using random forest and LASSO to predict wine quality

Exploring potentials of indigenous yeasts in fermentation of Chambourcin Grapes

Background: Indigenous yeasts present on grape berries have been shown to impact the winemaking process and final wine quality. In this study, we used Chambourcin, a hybrid grape cultivar, to isolate indigenous yeasts for potential use in winemaking. Hybrid grapes are of particularly interest due to higher resistance to cold temperature and fungal diseases. Methods: Yeasts were isolated from spontaneous fermentation of crushed grapes by plating on Dichloran Rose Bengal Chloramphenicol agar and identified by Sanger sequencing. Yeast candidates were selected for their ability to grow in Yeast Extract-Peptone-Dextrose broth supplemented with varying ethanol concentration. Candidate yeasts were chosen for mock fermentation using sterile Chambourcin juice. Volatile and non-volatile compounds that predict wine quality (flavor and aroma) were measured by UPLC and GC-MS. Results: Hanseniaspora uvarum, Starmerella bacillaris, Candida californica, and Zygoascus meyerae were predominantly isolated from Chambourcin. S. bacillaris isolate 180002 and C. californica isolate 180004 were able to tolerate up to 10% ethanol compared to S. cerevisiae (ethanol tolerance up to 12%). Our results demonstrate that isolate 180002 grown in ethanol supplemented medium is comparable to the commercial S. cerevisiae. Various aroma profiles of three main chemical families – esters, higher alcohol and volatile acids in products fermented by S. bacillaris and other isolates were observed. Conclusion: Starmerella bacillaris isolate 180002 demonstrates potential for use in winemaking with hybrid grapes based on ethanol tolerance and production of wine related chemical compounds. This study further supports the application of indigenous grape-associated yeasts in creating flavor and aroma diversity of wine.

A Model for Predicting Wine Quality in a Rhône Valley Vineyard

Core Ideas This modeling study to evaluate wine quality was based on a highly extensive and unique dataset. The model was validated with the actual scores attributed by the international press. This work can serve as a basis for studies in other vineyards. A model was developed as a tool to evaluate wine quality via submodels that represent soil, plants, meteorology, agronomic variables, various vineyard parameters, grape (Vitis vinifera L.) characteristics, and enological practices. The model is based on a highly extensive and unique dataset collected from a single Rhône Valley vineyard (Domaine de Beaurenard) in southern France and the expertise of its winemaker. This modeling study is a first step toward simulating the quality of a wine on the basis of 88 variables in the vineyard and 56 enological variables. The model was validated by comparing the scores predicted by the model with actual scores given by the international press for 49 modeled wines. At the end of the study, the model was compared against other statistical models. Through this model, winemakers can predict future wine quality early and develop improved business strategies. This modeling study was created for and applied directly to the studied situation (Châteauneuf du Pape, Côtes du Rhône, and Rasteau) and is not directly transferable to other situations. Nonetheless, it can serve as a basis for studies in other vineyards because, with the development of precision farming and the multiplication of sensors in agronomy, most wineries will soon have access to the necessary data to model the quality of their wine.

Selection of important features and predicting wine quality using machine learning techniques

Nowadays, industries are using product quality certifications to promote their products. This is a time taking process and requires the assessment given by human experts which makes this process very expensive. This paper explores the usage of machine learning techniques such as linear regression, neural network and support vector machine for product quality in two ways. Firstly, determine the dependency of target variable on independent variables and secondly, predicting the value of target variable. In this paper, linear regression is used to determine the dependency of target variable on independent variables. On the basis of computed dependency, important variables are selected those make significant impact on dependent variable. Further, neural network and support vector machine are used to predict the values of dependent variable. All the experiments are performed on Red Wine and White Wine datasets. This paper proves that the better prediction can be made if selected features (variables) are being considered rather than considering all the features.

The rise of phenology with climate change: an evaluation of IJB publications.

In recent decades, phenology has become an important tool by which to measure both the impact of climate change on ecosystems and the feedback of ecosystems to the climate system. However, there has been little attempt to date to systematically quantify the increase in the number of scientific publications with a focus on phenology and climate change. In order to partially address this issue, we examined the number of articles (original papers, reviews and short communications) containing the terms 'phenology' and 'climate change' in the title, abstract or keywords, published in the International Journal of Biometeorology in the 60years since its inception in 1957. We manually inspected all issues prior to 1987 for the search terms and subsequently used the search facility on the Web of Science online database. The overall number of articles published per decade remained relatively constant (255-378) but rose rapidly to 1053 in the most recent decade (2007-2016), accompanied by an increase (41-172) in the number of articles containing the search terms. A number of factors may have contributed to this rise, including the recognition of the value of phenology as an indicator of climate change and the initiation in 2010 of a series of conferences focusing on phenology which subsequently led to two special issues of the journal. The word 'phenology' was in use from the first issue, whereas 'climate change' only emerged in 1987 and peaked in 2014. New technologies such as satellite remote sensing and the internet led to an expansion of and greater access to a growing reservoir of phenological information. The application of phenological data included determining the impact of warming of phenophases, predicting wine quality and the pollen season, demonstrating the potential for mismatch to occur and both reconstructing and forecasting climate. Even though this analysis was limited to one journal, it is likely to be indicative of a similar trend across other scientific publications.

The Classification of White Wine and Red Wine According to Their Physicochemical Qualities

The main purpose of this study is to predict wine quality based on physicochemical data. In this study, two large separate data sets which were taken from UC Irvine Machine Learning Repository were used. These data sets contain 1599 instances for red wine and 4898 instances for white wine with 11 features of physicochemical data such as alcohol, chlorides, density, total sulfur dioxide, free sulfur dioxide, residual sugar, and pH. First, the instances were successfully classified as red wine and white wine with the accuracy of 99.5229% by using Random Forests Algorithm. Then, the following three different data mining algorithms were used to classify the quality of both red wine and white wine: k-nearest-neighbourhood, random forests and support vector machines. There are 6 quality classes of red wine and 7 quality classes of white wine. The most successful classification was obtained by using Random Forests Algorithm. In this study, it is also observed that the use of principal component analysis in the feature selection increases the success rate of classification in Random Forests Algorithm.

Evaluation of Wines Rating Based on Sensory Characteristics Using Neural Networks

Wine is an agricultural product with very high commerce price variation, which is highly affected by quality ratings. Therefore, quality rating is particularly important for both industry and consumers. However, absence of clear concepts on what constitutes wine quality makes the perception of quality highly subjective, and it is usual for tasters to disagree on the quality rating of a specific wine. For this purpose, a Feedforward Neural Network (FNN) could be trained in order to predict wine quality. In this study, a new FNN method was developed to predict the accurate wine quality based on major sensory characteristics as FNN inputs, and to improve the ability of a taster, groups of tasters, or consumers, to rate wine by taking into account previous decisions. Specifically, five principle sensory characteristics of wines were used as inputs (Aging in Barrel, Aroma Intensity, Body, Astringency, and Acidity) in a rating range 1-3. As outputs, the quality ratings of wines in a range 70-100 were considered. The FNN was created in MATLAB with 1 hidden layer, 5 neurons and 1 output layer. For ratings divided in 5 categories the accuracy was 53% with the use of the FNN, as opposed to the accuracy of 36% achieved by Multiple Linear Regression. For ratings divided in 9 categories the accuracy was 90%. This method may allow each individual or group of tasters to introduce their own data to produce a more objective rating by taking into account previous decisions (subjective) that have accumulated in the database.