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…

Terroir et marché : exemples de stratégie pour les vins d’une petite région (Muscadet – Anjou – Touraine)

The designations of origin of the Loire Valley wine have been recognized according to customs and notoriety established over the centuries since the Middle Ages. There are four main production basins going up the Loire, from Nantes to the Sancerrois region: Nantes, Anjou-Saumur, Touraine and the vineyards of the Centre. In each of these basins, there is a wide range of appellations of origin which has been established according to a logic which may not seem obvious to the uninformed.

Advances in the chemistry of rosé winemaking and ageing

The market share of Rosé wine in France has grown from 11 % to 32 % over the last 20 years. Current trends are towards rosé wines of a lighter shade of pink, and where possible, containing a greater concentration in varietal thiols. Grape varieties, the soil on which they are grown, viticultural practices and winemaking technology all impact the polyphenols, color and aromas of rosé wines.

Three proximal sensors to estimate texture, skeleton and soil water storage in vineyards

Proximal sensors are becoming widely used in precision viticulture, due to the quick, easy and non-invasive identification of soil spatial variability. The apparent soil electrical conductivity (ECa) is the main parameter measured by sensors, which is correlated to many factors, like soil water content, salinity, clay content and mineralogy, rock fragments, bulk density, and porosity.

Effect on the grape and wine characteristics of cv. Tempranillo at 3 production levels

The vineyard has experienced a general increase in yields mainly due to the elevated use of technology which caused a quality loss of grapes in more than one case. A large percentage of the Spanish vineyard is covered by a Denomination of Origin which limits the productive level of the vineyards as one of its regulations. The maximum production limit is a variable characteristic of each vineyard and is not usually regulated by agronomic criteria, and this explains the fact that each vineyard can reach high quality with a totally different yield from that set by the Denomination of Origin.

Impact of non-fruity compounds on red wines fruity aromatic expression: the role of higher alcohols

A part, at least, of the fruity aroma of red wines is the consequence of perceptive interactions between various aromatic compounds, particularly ethyl esters and acetates, which may contribute to the perception of fruity aromas, specifically thanks to synergistic effects.1,2 The question of the indirect impact of non-fruity compounds on this particular aromatic expression has not yet been widely investigated. Among these compounds higher alcohols (HA) represent the main group, from a quantitative standpoint, of volatiles in many alcoholic beverages. Moreover, some bibliographic data suggested their contribution to the aromatic complexity by either increasing or masking flavors of wine, depending of their concentrations.

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…

A Viticultural Terroir in Brazil: Change and continuity

The viticultural terroir at the Serra Gaúcha region, in Rio Grande do Sul State, Brazil, is analyzed under historical and sociological viewpoints, aiming to understand the origin of its characteristics, and the risks for its continuity.

Control of grapevine virus diseases in collections and at the stages of propagation in Ukraine

The principles of virological control on different types of grapevine collections and plantations are summarized.

Caractérisation du terroir en Espagne : méthodologie de l’évaluation et de la validation

In recent years, there has been a growing interest in characterizing the ecological environment of vineyard production, and the growing need to delimit and characterize with precision the different homogeneous viticultural units. This has allowed the development of new studies which have as their objective the Vineyard Zoning. The delimitation and characterization of wine-growing areas poses specific problems in Spain, not only linked to the specific characteristics of the territory, but also to the size, distribution and index of viticultural occupation in the designations of origin.

Rootstock differences in soil-water uptake during drying-wetting cycles imaged with 3d electrical resistivity tomography

Limited knowledge has been acquired on grapevine roots and rhizosphere processes because of harder access when compared to aerial parts. There is need for new methods to study root behavior in undisturbed field conditions, and relate these effects on canopy and yield. The aim of this multidisciplinary study was to image and quantify spatial-temporal differences in soil-water uptake by genetically different rootstocks and to assess the response of the canopy during drought and rewetting.

Rootstock drought tolerance under dry-farmed conditions in Oregon’s Willamette Valley

Rootstocks are used in vineyards worldwide and have been the focus of many studies. However, rootstock performance varies based on regional climates and soil types. As Oregon experiences warmer seasons and variable precipitation patterns, growers are interested in rootstocks with more drought tolerance than the commonly planted rootstocks: 3309C, Riparia Gloire, and 101-14 Mgt. In Oregon’s Willamette Valley, annual precipitation is typically sufficient to make dry-farming possible and use of irrigation is limited.