Modeling wine preferences by data mining from physicochemical properties
By Paulo Cortez, António Cerdeira, Fernando Almeida, Telmo Matos, José Reis. 2009
1,519 citations recorded by the source corpus.
We propose a data mining approach to predict human wine taste preferences that is based on easily available analytical tests at the certification step. A large dataset (when compared to other studies in this domain) is considered, with white and red vinho verde samples (from Portugal). Three regression techniques were applied, under a computationally efficient procedure that performs simultaneous variable and model selection. The support vector machine achieved promising results, outperforming the multiple regression and neural network methods. Such model is useful to support the oenologist wine tasting evaluations and improve wine production. Furthermore, similar techniques can help in target marketing by modeling consumer tastes from niche markets.
Subfields: Food Science, Plant Science, Tourism, Leisure and Hospitality Management
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Food Science
In Plant Science
In Tourism, Leisure and Hospitality Management
Citations in the corpus
- Privacy-Preserving Distributed Linear Regression on High-Dimensional Data (2017, neutral): “1 · 103 24 53 4 Wine Quality [12, 13] 11 4898 2.”
- Using Machine Teaching to Identify Optimal Training-Set Attacks on Machine Learners (2015, methodology): “This data set consists of n = 4601 instances, each of which has d = 57 numerical features x and a binary label yi (1 for spam and −1 for not spam).”
- Opening black box Data Mining models using Sensitivity Analysis (2011, methodology): “To demonstrate the visualization of input importances, we selected the wwq-reg dataset, which includes 11 inputs.”
- Speeding Up Distributed Machine Learning Using Codes (2015, methodology): “We use wine quality data set for linear regression, and Spambase data set for classification, both from UCI Machine Learning Repository [1,5].”
- A data-driven approach to predict the success of bank telemarketing (2014, neutral): “For instance, SVM provided better results in [6][8], comparable NN and SVM performances were obtained in [5], while DT outperformed NN and”
- The Classification of White Wine and Red Wine According to Their Physicochemical Qualities (2016, background): “Wine industry is researching new technologies for both wine making and selling processes in order to back up this growth [1].”
- Learning Decision Trees with Flexible Constraints and Objectives Using Integer Optimization (2017, neutral): “The first one “RedWine” is from the wine quality dataset [7].”
- PrivLogit: Efficient Privacy-preserving Logistic Regression by Tailoring Numerical Optimizers (2016, methodology): “Among these, we have compiled four real-world studies, including: 1) the Wine quality study (with 6,497 samples and 12 features) [Cortez et al., 2009] for predicting wine quality from physicochemical tests, 2) online Loans data (with 122, 578 samples and 33 features) from Lending Club [LengingClub,…”
- Autotune: A Derivative-free Optimization Framework for Hyperparameter Tuning (2018, background): “The data set is a collection of red and white variants of the Portuguese “Vino Verde” wine [2], with 6,497,000 observations and 11 features representing the physiochemical properties of the wines and a quality rating for each wine.”
- Convex Formulations for Fair Principal Component Analysis (2018, methodology): “Next, we consider the sensitivity of our results to these parameters for the Wine Quality dataset (Cortez et al., 2009).”
- Using sensitivity analysis and visualization techniques to open black box data mining models (2013, neutral): “Email addresses: pcortez@dsi.uminho.pt (Paulo Cortez), embrem@rpi.edu (Mark J. Embrechts) Preprint submitted to Information Sciences October 12, 2012”
- Data Mining with Neural Networks and Support Vector Machines Using the R/rminer Tool (2010, background): “Also, it is possible to extract knowledge from NNs and SVMs, given in terms of input relevance [4].”
- Bayesian and Empirical Bayesian Forests (2015, background): “The first example, taken from (Cortez et al., 1998), involves prediction of an ‘expert’ quality ranking on the scale of 0-10 for wine based upon 11 continuous attributes (physiochemical properties of the wine) plus wine color (red or white).”
- General Latent Feature Models for Heterogeneous Datasets (2017, background): “Wine quality dataset [6] 6,497 12 (11 P + 1 N) Results of physicochemical tests realized to different wines.”