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…

Grapevine drought tolerant ideotypes to adapt viticulture to climate change

Climate change is challenging the resilience of grapevine, one of the most important crops worldwide. Adapting viticulture to a hotter and drier future will require a multifaceted approach that must include new management strategies, increased irrigation efficiency, and the identification of more drought tolerant genotypes.

EVALUATING WINEMAKING APPLICATIONS OF ULTRAFILTRATION TECHNOLOGY

Ultrafiltration is a process that fractionates mixtures using semipermeable membranes, primarily on the basis of molecular weight. Depending on the nominal molecular weight cut-off (MWCO) specifications of the membrane, smaller molecules pass through the membrane into the ‘permeate’, while larger molecules are retained and concentrated in the ‘retentate’. This study investigated applications of ultrafiltration technology for enhanced wine quality and profitability. The key objective was to establish to what extent ultrafiltration could be used to manage phenolic compounds (associated with astringency or bitterness) and proteins (associated with haze formation) in white wine.

Factors influencing cover crop water competition in vineyards and implications for future drought adaptation

Vineyard water management in Australia is often associated with irrigation in warm and hot climates, but in cooler regions the larger share of the seasonal water demand is met by rainfall.

The effect of sulfur compounds on the formation of varietal thiols in Sauvignon Blanc and Istrian Malvasia wines

Varietal thiols 3-sulfanylhexan-1-ol (3SH), 3-sulfanylhexyl acetate (3SHA) and 4-methyl-4-sulfanylpentan-2-one (4SMP) are essential for fruity aromas of Sauvignon Blanc wines. The concentration of varietal thiols in wines was thought to be related to the concentration of their precursors in grapes, however only a small proportion of precursors are released to varietal thiols during fermentation. New findings suggested that specific grape juice metabolites could significantly impact on the development of three major varietal thiols and other aroma compounds of Sauvignon Blanc wines.

Impact of geographical location on the phenolic profile of minority varieties grown in Spain. II: red grapevines

Because terroir and cultivar are drivers of wine quality, is essential to investigate theirs effects on polyphenolic profile before promoting the implantation of a red minority variety in a specific area. This work, included in MINORVIN project, focuses in the polyphenolic profile of 7 red grapevines minority varieties of Vitis vinifera L. (Morate, Sanguina, Santafe, Terriza Tinta Jeromo Tortozona Tinta) and Tempranillo) from six typical viticulture Spanish areas: Aragón (A1), Cataluña (A2), Castilla la Mancha (A3), Castilla –León (A4), Madrid (A5) and Navarra (A6) of 2020 season. Polyphenolic substances were extracted from grapes. 35 compounds were identified and quantified (mg subtance/kg fresh berry) by HPLC and grouped in anthocyanins (ANT) flavanols (FLAVA), flavonols (FLAVO), hydroxycinnamic (AH), benzoic (BA) acids and stilbenes (ST). Antioxidant activity (AA, mmol TE /g fresh berry) was determined by DPPH method. The results were submitted to a two-way ANOVA to investigate the influence of variety, area and their interaction for each polyphenolic family and cluster analysis was used to construct hierarchical dendrograms, searching the natural groupings among the samples. Sanguina (A3) had the most of total polyphenols while Tempranillo (A5) those of ANT. Sanguina (A2) and (A3) reached the highest values of FLAVO, FLAVA and AA. These two last samples had also the maximum of AA. The effect cultivar and area were significant for all polyphenolic families analyzed. A high variability due to variety (>50%) was observed in FLAVA and the maximum value of variability due to growing area was detected in AA (86.41%), ANT and FLAVO (51%); the interaction variety*zone was significant only for ANT, FLAVO, EST and AA. Finally, dendrograms presented five cluster: i) Sanguina (A2); ii) Sanguina (A3); iii) Tempranillo (A5); iv) Tempranillo (A3); Terriza (A3,A5), Morate (A5,A6); v) Santafé (A1,A6); Tortozona tinta (A1,A3,A6); Tinta Jeromo (A3,A4).

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…

Exploring changes in browning kinetics, color, and antioxidants due to dealcoholization of wine

The global consumer demand for low or non-alcoholic wine is growing steadily in recent years, driven by health concerns, religious beliefs, and personal taste preferences etc.. Consequently, the removal of alcohol from wine can significantly alter its chemical and sensory properties, including color, aroma, and taste, which make a significant challenge for consumer to accept these products. Ethanol plays a crucial role in various chemical reactions and interactions that contribute to the development of wine’s characteristics.

Mannoprotein extracts from wine lees: characterization and impact on wine properties

This study aims at exploiting an undervalued winemaking by-product, wine yeast lees, by developing efficient and food-grade methods for the extraction of yeast glycoproteins. These extracts were then supplemented to wine and their impact on wine properties assessed.

Impact of Ecklonia maxima seaweed extract on the vegetative, reproductive and microbiome in Vitis vinifera L. cv Cabernet-Sauvignon

Context and purpose of the study. Climate change is a major challenge in wine production. It results in erratic weather conditions which may lead to a reduction in grape yield and the subsequent grape and wine quality.

Symbiotic microorganisms application in vineyards: impacts on grapevine performance and microbiome

Microorganism-based inoculants have been suggested as a viable solution to mitigate the adverse effects of climate change on viticulture. However, the actual effectiveness of these inoculants when applied under field conditions remains a challenge, and their effects on the existing soil microbiota are still uncertain. This study investigates the impact of arbuscular mycorrhizal fungi inoculation on grapevine performance and microbiome. The study was conducted in a vineyard of Callet cultivar in Binissalem, Mallorca, Spain. Two different treatments were applied: control and inoculation with commercial mycorrhizae complex of Rhizoglomus irregulare applied to plants through irrigation.

Rationalising the impact of time, light, temperature, and oxygen on the evolution of rosé wines by means of a surface response methodology approach

The widespread use of flint glass bottles for rosé wines is driven by consumer preference for color as a key choice factor.