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…

Evaluation of climate change impacts at the Portuguese Dão terroir over the last decades: observed effects on bioclimatic indices and grapevine phenology

In the last decades the growers of the Portuguese Dão winegrowing region (center of Portugal) are experiencing changes in climate that are influencing either grape phenology berry health and ripening. Aiming to study the relationships between climate indices (CI), seasonal weather and grapevine phenology, in this work long-term climate and phenological data collected at the experimental vineyard of the Portuguese Dão research centre between 1958 and 2019 (61 years) for the red variety Touriga Nacional, was analyzed. The trends over time for the classical temperature-based indices (Growing Season Temperature – GST -, Growing Degree Days – GDD, Huglin Index – HI and Cool Night Index – CI) presented a significantly positive slope while the Dryness Index (DI) showed a negative trend over the last 61 years. Regarding grapevine phenology, an average advance of 4.5 days per decade in the harvest day was observed throughout the last 61 years. Consequently, the weather conditions during the ripening period have changed, showing an increasing trend over time in the average temperature (higher magnitude in the maximum than in the minimum temperature) and a decrease in the accumulated rainfall. A regression analysis showed that ~50% of harvest date variability over years was explained by the temperature-based indices variability. These observed effects of climate change on bioclimatic indices and corresponding anticipation of harvest date can still be considered advantageous for the Dão terroir as it allows to achieve an optimal berry ripening before the common equinox rains and, therefore, avoid the potential negative impacts of the rainfall on berry health and composition.

Colored hail‐nets as a tool to improve vine water status: effects on leaf gas exchange and berry quality in Italia table grape

Protecting table grape vineyards with white hail‐nets is a common practice in Southern Italy. Hail‐nets result in shading effects of 10‐20 %, depending on their density

Impact of industrial-scale serial filtration on macromolecules in red wines

Filtration is a critical step in ensuring the clarity and microbial stability of wine prior to bottling. However the process of filtering potentially reduces red wine quality by removing some of the macromolecules that contribute to the texture of the wine. Commercial red wines, Cabernet Sauvignon (CAS) and Shiraz (SHZ), of two vintages and two grades (premium grade wines from the older vintage: CAS13 and SHZ13; and standard grade wines from a younger vintage: CAS14 and SHZ14) were filtered through industrial-scale commercial filtration units prior to bottling. Samples were taken before and after cross-flow filtration, lenticular filters, 0.65 µm and 0.45 µm pore size nylon membrane filters. The concentration and composition of macromolecules, including tannins and polysaccharides, were measured in all samples as well as particle size distribution and wine colour.

Exploring the prevalence of esca-induced leaf symptoms in French vineyards and the role of climate: a national scale analysis

Esca, a severe trunk disease affecting vineyards, is caused by fungal pathogens that induce wood necrosis and decay, leaf symptoms, yield losses, and potentially a rapid death of the vine. The prevalence of this disease varies across years, regions, cultivars, and plot ages. Despite its significance in understanding and predicting dieback risk in different vineyards, the role of climate in trunk diseases remains a relatively unexplored research area. While some studies have demonstrated the impact of certain climatic conditions on the prevalence of the disease, they often focus on a limited number of plots and yield conflicting results.We conducted a statistical analysis, using a Bayesian approach on a national database comprising prevalence data of esca from over 500 different plots in France, spanning the years 2003 to 2022 and encompassing various cultivars.

Effect of mannoproteins extracted from Torulaspora delbrueckii on wine flavanol composition and on flavanol-salivary protein interactions

Global climate change is exerting an influence on vine phenology, leading to a decoupling of technological and phenolic maturity of grapes. This results in the modification of berry chemical composition, which can translate into wines with excessive astringency. The addition of mannoproteins (MP) to wine has been proposed as a way of mitigating this problem, since some studies have shown that MPs can modulate wine astringency. However, the mechanism underlying the astringency modulation effect of MPs is not well known and it seems to be dependent on the compositional and structural characteristics of the MP.

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…

Rare earth elements in grapes and soil: study of different soil extraction methods

Lanthanides, together with scandium and yttrium, make up the group of Rare Earth Elements (REEs). An official method for analysis of the bioavailable REEs accumulated by plants, depending mainly on soil characteristics, chemical speciation in soil and the specific ability of the plant, is still lacking.

Water deficit differentially impacts the performances and the accumulation of grape metabolites of new varieties tolerant to fungi

The use of resistant varieties is a long-term but promising solution to reduce chemical input in viticulture. Several important breeding programs in Europe and abroad are now releasing a range of new hybrids performing well regarding fungi susceptibility and producing good quality wines. Unfortunately, insufficient attention is paid by the breeders to the adaptation of these varieties to climatic changes, notably to the increased climatic demand and water deficit (WD). Thus, prior to the adoption of such varieties by the wine industry in Mediterranean regions, there is a need to consider their suitability to WD. This study aimed to characterize the different drought-strategies adopted by 6 new resistant varieties selected by INRAE in comparison to Syrah. To allow the assessment of long-term impacts of WD, field-grown vines were exposed to contrasted WD from 2018 to 2021 under a semi-arid Mediterranean climate. A gradient of WD was applied in the field and controlled through plant measurements at the single plant level. Grape development was non-destructively monitored to determine the arrest of berry phloem unloading. The impacts of WD on berry composition, including water, primary metabolites (sugars, organic acids), secondary metabolites (anthocyanins, thiols precursors) and main cations contents, were assessed at this specific stage. Results showed different varietal responses during the year and inter-annual acclimation in terms of plant water use efficiency, biomass accumulation, as well as yield components and berry composition. WD differentially reduced the accumulation of primary metabolites at plant and berry levels, but it little changed their concentrations in the fruits at the ripe stage. Moreover, WD differentially impacted the accumulation of secondary metabolites and major cations between the varieties. In the talk, we’ll present the main results regarding the WD impacts on fruit metabolites and enlarge the reflection about the practical assessment of the grapevine acclimation to WD.

Effects of winemaking variables on the chemical and sensory quality of Schiava wines up to one year storage in bottle

The interactive effects of three major enological variables were evaluated on the quality of Schiava wine up to one year of storage in bottle.

Variability in the content of coarse elements in a viticultural plot in the Graves appellation: relationship with geophysical data

Il a été souvent démontré (Seguin, 1970), que les meilleurs terroirs sont ceux qui présentent pendant la période de maturation du raisin, une régulation et une limitation de l’alimentation hydrique de la vigne. Si on s’intéresse aux facteurs influençant ce régime hydrique, on constate le rôle prépondérant du taux d’éléments grossiers non poreux qui limitent la réserve utile du sol en diminuant le taux de terre fine. De plus, ces éléments grossiers jouent également un rôle au niveau du pédo-climat thermique car leur conductivité thermique et leur chaleur spécifique sont plus élevées que celles de la terre fine. Ainsi le sol se réchauffera et se refroidira plus rapidement (Saini et McLean, 1967), (Gras, 1994).

Design of an indicator of vine vigor potential conferred by soil (vipos), using a fuzzy expert system

Winegrowers must adapt more and more their viticultural practices in order to evolve toward a sustainable viticulture, to be competitive and to improve both the production methods and the quality and typicalness of wines. In this context, ‘Terroir’ studies in Loire Valley vineyards have allowed to build decision aid maps that can be used directly by growers to adjust their practices.