Finding the most important sensory descriptors to differentiate some Vitis vinifera L. South American wines using support vector machines

  • Nattane Luíza Costa
  • Laura Andrea García Llobodanin
  • Inar Alves Castro
  • Rommel BarbosaEmail author


The geographical recognition of wines has been extensively attempted based on chemical parameters. However, few studies have used wine sensory properties to characterize wines according to their geographical origin. This paper presents a machine learning study to classify and to find the most important sensory descriptors of Cabernet Sauvignon, Syrah, Tannat, and Merlot wines from Argentina, Brazil, Chile and Uruguay. Four feature selection methods (F score, relief, \({\chi ^2}\), and random forest importance) were used to generate the order of importance of the sensory descriptors. The feature subsets were generated based on the feature selection ranking order to use as input features for the support vector machines classifier. Very good results with 85–100% accuracy were achieved, and the results showed that few sensory descriptors discriminate the origin of wines better than when using all the descriptors and that there is a specific subset of most important features to each wine variety. As far as we know, this is the first study to analyze South American wines based solely on sensory descriptors and support vector machines along with feature selection methods.


Machine learning South American wines Sensory descriptors Classification Support vector machines Feature selection 



We would like to acknowledge Fapesp (Fundação de Amparo à Pesquisa do Estado de São Paulo) for their financial support and the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Brasil (CAPES).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no competing interests.

Compliance with ethics requirements

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee.


  1. 1.
    Van Leeuwen C, Seguin G (2006) The concept of terroir in viticulture. J Wine Res 17:1–10CrossRefGoogle Scholar
  2. 2.
    Versari A, Laurie VF, Ricci A et al (2014) Progress in authentication, typification and traceability of grapes and wines by chemometric approaches. Food Res Int 60:2–18CrossRefGoogle Scholar
  3. 3.
    Lesschaeve I (2007) Sensory evaluation of wine and commercial realities: review of current practices and perspectives. Am J Enol Vitic 58:252–258Google Scholar
  4. 4.
    Geana I, Iordache A, Ionete R et al (2013) Geographical origin identification of Romanian wines by ICP-MS elemental analysis. Food Chem 138:1125–1134CrossRefGoogle Scholar
  5. 5.
    Falqué E, Darriet P, Fernández E, Dubourdieu D (2008) Volatile profile and differentiation between Albariño wines from different origins. Int J Food Sci Technol 43:464–475CrossRefGoogle Scholar
  6. 6.
    Azcarate SM, de Araújo Gomes A, Alcaraz MR et al (2015) Modeling excitation–emission fluorescence matrices with pattern recognition algorithms for classification of Argentine white wines according grape variety. Food Chem 184:214–219CrossRefGoogle Scholar
  7. 7.
    Fanzone M, Zamora F, Jofré V et al (2012) Phenolic characterisation of red wines from different grape varieties cultivated in Mendoza province (Argentina). J Sci Food Agric 92:704–718CrossRefGoogle Scholar
  8. 8.
    King ES, Stoumen M, Buscema F et al (2014) Regional sensory and chemical characteristics of Malbec wines from Mendoza and California. Food Chem 143:256–267CrossRefGoogle Scholar
  9. 9.
    Green JA, Parr WV, Breitmeyer J et al (2011) Sensory and chemical characterisation of Sauvignon blanc wine: influence of source of origin. Food Res Int 44:2788–2797CrossRefGoogle Scholar
  10. 10.
    Kallithraka S, Arvanitoyannis IS, Kefalas P et al (2001) Instrumental and sensory analysis of Greek wines; implementation of principal component analysis (PCA) for classification according to geographical origin. Food Chem 73:501–514CrossRefGoogle Scholar
  11. 11.
    Granato D, Uchida Katayama FC, de Castro IA (2012) Characterization of red wines from South America based on sensory properties and antioxidant activity. J Sci Food Agric 92:526–533CrossRefGoogle Scholar
  12. 12.
    Llobodanin LG, Barroso LP, Castro IA (2014) Sensory characterization of young South American red wines classified by varietal and origin. J Food Sci 79:S1595–S1603CrossRefGoogle Scholar
  13. 13.
    García-Muñoz S, Muñoz-Organero G, Fernández-Fernández E, Cabello F (2014) Sensory characterisation and factors influencing quality of wines made from 18 minor varieties (Vitis vinifera L.). Food Qual Prefer 32:241–252CrossRefGoogle Scholar
  14. 14.
    Zielinski AAF, Haminiuk CWI, Nunes CA et al (2014) Chemical composition, sensory properties, provenance, and bioactivity of fruit juices as assessed by chemometrics: a critical review and guideline. Compr Rev food Sci food Saf 13:300–316CrossRefGoogle Scholar
  15. 15.
    Maione C, Barbosa RM (2018) Recent applications of multivariate data analysis methods in the authentication of rice and the most analyzed parameters: a review. Crit Rev Food Sci Nutr 1–12Google Scholar
  16. 16.
    Cortez P, Cerdeira A, Almeida F et al (2009) Modeling wine preferences by data mining from physicochemical properties. Decis Support Syst 47:547–553CrossRefGoogle Scholar
  17. 17.
    da Costa NL, Castro IA, Barbosa R (2016) Classification of cabernet sauvignon from two different countries in South America by chemical compounds and support vector machines. Appl Artif Intell 30:679–689CrossRefGoogle Scholar
  18. 18.
    Gómez-Meire S, Campos C, Falqué E et al (2014) Assuring the authenticity of northwest Spain white wine varieties using machine learning techniques. Food Res Int 60:230–240CrossRefGoogle Scholar
  19. 19.
    da Costa NL, Llobodanin LAG, de Lima MD et al (2018) Geographical recognition of Syrah wines by combining feature selection with extreme learning machine. Measurement 120:92–99CrossRefGoogle Scholar
  20. 20.
    Chandrashekar G, Sahin F (2014) A survey on feature selection methods. Comput Electr Eng 40:16–28CrossRefGoogle Scholar
  21. 21.
    Chen Y-W, Lin C-J (2006) Combining SVMs with various feature selection strategies. In: Feature extraction. Springer, New York, pp 315–324Google Scholar
  22. 22.
    Kenji, Rendell LAK (1992) The feature selection problem: traditional methods and a new algorithm. AAAI 2:129–134Google Scholar
  23. 23.
    Tang J, Alelyani S, Liu H (2014) Feature selection for classification: a review. Data Classif Algorithms Appl 37–64Google Scholar
  24. 24.
    Robnik-Šikonja M, Kononenko I (2003) Theoretical and empirical analysis of ReliefF and RReliefF. Mach Learn 53:23–69CrossRefGoogle Scholar
  25. 25.
    Metin SK (2018) Feature selection in multiword expression recognition. Expert Syst Appl 92:106–123CrossRefGoogle Scholar
  26. 26.
    Breiman L (2001) Random forests. Mach Learn 45:5–32CrossRefGoogle Scholar
  27. 27.
    Cilia ND, De Stefano C, Fontanella F, di Freca AS (2018) A ranking-based feature selection approach for handwritten character recognition. Pattern Recognit Lett. Google Scholar
  28. 28.
    Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20:273–297Google Scholar
  29. 29.
    Xue H, Yang Q, Chen S (2009) SVM: support vector machines. In: Wu X, Kumar V (eds) The topten algorithms in data mining, 1st ed, Chapman and Hall/CRC, pp 37–59Google Scholar
  30. 30.
    Jurado JM, Alcázar Á, Palacios-Morillo A, De Pablos F (2012) Classification of Spanish DO white wines according to their elemental profile by means of support vector machines. Food Chem 135:898–903CrossRefGoogle Scholar
  31. 31.
    R Core Team (2018) R: a language and environment for statistical computing. Accessed 15 Mar 2018
  32. 32.
    Romanski P, Kotthoff L, Kotthoff ML (2016) Package FSelector: selecting attributes. Accessed 15 Mar 2018
  33. 33.
    Kuhn M, Contributions from Wing J, Weston S et al (2017) caret: classification and regression training. Accessed 15 Mar 2018
  34. 34.
    Wickham H (2009) ggplot2: elegant graphics for data analysis. Springer, New YorkCrossRefGoogle Scholar
  35. 35.
    Meyer D, Dimitriadou E, Hornik K et al (2017) e1071: misc functions of the Department of Statistics, Probability Theory Group (formerly: E1071), TU Wien. Accessed 15 Mar 2018
  36. 36.
    Anzanello MJ, Fogliatto FS, Rossini K (2011) Data mining-based method for identifying discriminant attributes in sensory profiling. Food Qual Prefer 22:139–148CrossRefGoogle Scholar
  37. 37.
    Yu P, Low MY, Zhou W (2018) Design of experiments and regression modelling in food flavour and sensory analysis: a review. Trends Food Sci Technol 71:202–215CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Instituto de InformáticaUniversidade Federal de GoiásGoiâniaBrazil
  2. 2.Department of Food and Experimental Nutrition, Faculty of Pharmaceutical Sciences, LADAFUniversity of São PauloSão PauloBrazil

Personalised recommendations