Macrowine 2021
IVES 9 IVES Conference Series 9 Beyond classical statistics – data fusion coupled with pattern recognition

Beyond classical statistics – data fusion coupled with pattern recognition

Abstract

AIM: Patterns in data obtained from wine chemical and sensory evaluations are difficult to infer using classical statistics. Pattern recognition can be resolved by coupling data fusion with machine learning techniques, possibly leading to new hypotheses being formed. This study demonstrates the applicability of two pattern recognition approaches using as case study involving Chenin Blanc wines (recently bottled and after two years storage) from young (35 years) vines.

METHODS: Sensory (sorting (Mafata et al. 2020)) and chemical (NMR: nuclear magnetic resonance, HRMS: high resolution mass spectrometry, and UV-Vis: ultraviolet spectrophotometry) data were collected for the young and aged (two years in the bottle) wines. Data sets were combined using multiple factor analysis (MFA). Exploratory unsupervised cluster analysis was performed by agglomerative hierarchical clustering (AHC) and Fuzzy-k means (Bezdek 1981). Optimal cluster conditions were found for both methods and the cophenetic coefficient was used to assess the most confident clustering method.

RESULTS: Since large data sets were fused, the models were very complex. There were no consistent clustering patterns when varying clustering conditions, signalling high similarity between samples. The samples could not confidently be distinguished from one another even at the highest optimized conditions. Although Fuzzy-k means gave more confident clustering, it was still not sufficient for solving classification issues in this sample set.

CONCLUSIONS:

Fuzzy-k means was better at resolving the natural grouping of samples. Coupled to data fusion, it could potentially lead to better pattern recognition, especially for oenological chemical and sensory data. The fuzzy approach should be explored, keeping in mind it is more sensitive to small differences in the data compared to classical statistics.

DOI:

Publication date: September 7, 2021

Issue: Macrowine 2021

Type: Article

Authors

Mpho Mafata, Jeanne

1South African Grape and Wine Research Institute, Department of Viticulture and Oenology, Stellenbosch University & 2School for Data Science and Computational Thinking, Stellenbosch University, South Africa, BRAND, South African Grape and Wine Research Institute, Department of Viticulture and Oenology, Stellenbosch University, South Africa  Astrid, BUICA, South African Grape and Wine Research Institute, Department of Viticulture and Oenology, Stellenbosch University

Contact the author

Keywords

data fusion, pattern recognition, machine learning, artificial intelligence, multiple factor analysis, fuzzy-k means, cluster analysis

Citation

Related articles…

A tool for catching mice in wine: development and application of a method for the detection of mousy off-flavour compounds in wine

Over the past two years, the AWRI has received 69 wine samples suspected of being affected by mousy off-flavour. The character has been mostly observed in white wines.

Optimised extraction and preliminary characterisation of mannoproteins from non-Saccharomyces wine yeasts

The use of non-Saccharomyces yeast species for the improvement of wine technological and oenological properties is a topic that has gained much interest in recent years [1]. Their application as co-starter cultures sequential to the inoculation of Saccharomyces cerevisiae and in aging on the lees has been shown to improve aspects such as protein stability and mouthfeel [2].

Breeding grapevines for disease and low temperature tolerance: the U.S. perspective

Most grape scion cultivars grown around the world are derived from a single species, Vitis vinifera. Yet, the proportion of interspecific hybrids is increasing for a variety of reasons, including resistance to abiotic stresses such as low temperatures; societal, economic and environmental pressures to reduce pesticide usage; and to add a greater range of flavors to new table grape cultivars.

Influence of dehydration and maceration conditions on VOCs composition and olfactory profile of Moscato Bianco passito sweet wine

Among the Vitis vinifera L. cv. Moscato, Moscato Bianco is the oldest and most cultivated one in Europe (1). According to the OIV Focus 2015, Italy is the country with the largest cultivated area of Moscato Bianco with about 12500 hectares (2), that is used to produce well-known wines (i.e., Moscato Passito in Piedmont, Moscato di Trani in Puglia, and Moscatello di Montalcino in Tuscany), mainly obtained from partially dehydrated grapes (1). Different dehydration techniques can strongly modify the chemical compounds of oenological interest, among which Volatile Organic Compounds (VOCs) (1) that are the main responsible for the varietal sensory character of the final wine.

Influence of must fining on oxygen consumption rate, oxidation susceptibility and electrochemical characteristics of different white grape musts

AIM: Pre-fermentative fining is one of the central steps of white wine production. Mainly aiming at reducing the levels of suspended solids, juice fining can also assist in reducing the content of oxidizable phenolics and therefore the susceptibility of juice to oxidation.

Macrowine 2021
IVES 9 IVES Conference Series 9 Beyond classical statistics – data fusion coupled with pattern recognition

Beyond classical statistics – data fusion coupled with pattern recognition

Abstract

Content of the article

References

Section for all references

DOI:

Publication date: September 7, 2021

Issue: (ex: Issue: Terclim 2023)

Type: typeofpublication

Authors

author1, author2, author3

Presenting author

Description

List of affiliations ¹ ² ³

Contact the author

Email address (with mailto: link)

Keywords

List of different keywords (keyword1, keyword2, keyword3)

Tags

Citation

Related articles…

Outside and inside grapevine roots: arbuscular mycorrhizal fungal communities in a ‘nebbiolo’ vineyard 

In field conditions, grapevine roots are colonized by arbuscular mycorrhizal fungi (AMF). Little is known about the species composition of AMF communities associated to grapevine.

Aroma typicity of Timorasso wines: influence of ageing on volatile organic compounds and sensory descriptors

‘Timorasso’ is an autochthonous white grape variety from southern Piedmont (Italy) used for producing wines in the Colli Tortonesi product designation of origin (PDO). Over the last decade, there has been a notable rise in its production, due to the increased interest of wine enthusiasts who prized its wine distinctive ageing notes [1].

Radiative and thermal effects on fruit ripening induced by differences in soil colour

One of the intrinsic parts of a vineyard “terroir” is soil type and one of the characteristics of the soil is it’s colour. This can differ widely from bright white, as for some calcareous soils, to red, as in “terra rossa” soils, or black, as in slate soils.

Monitoring small-scale alcoholic fermentations using a portable FTIR-ATR spectrometer and multivariate analysis

Although some wine production processes still rely on post-production evaluation and off-site laboratory analysis, the new winemaking industry is aware of a need for a better knowledge of the process to improve the properties of the final product. Thus, more and more wineries are interested in incorporating quality-by-design (QbD) strategies instead of postproduction testing because of the possibility to early detect deviations in fermentation or any other wine process. This would allow to detect unwanted situations and eventually to ‘readjust’ the process, thus minimizing rejects.

Which heat test can realistically estimate white wine haze risk?

Different heat tests are used to predict the dose of bentonite necessary to prevent wine haze after bottling. The most used tests are 60-120 min. at 80°C. Nevertheless, there is a lack of information about the relationship between these tests and the turbidities observed in the bottles after the storage/transport of the wines in realistic conditions, when temperatures reach 35-42°C during 3-12 days.