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…

Riesling as a model to irrigate white wine grape varieties in arid climates

Regulated deficit irrigation (RDI) is a common viticultural practice for wine grape production. In addition to the potential improvement of water use efficiency, the adoption of this technique favors smaller canopies with higher levels of fruit sun exposure, enhancing quality attributes associated with red wine grapes such as smaller berries with higher tannins and anthocyanins. However, these quality attributes do not necessarily transfer to white wine grapes. The goal of this project was to assess whether partial rootzone drying (PRD) is more suited than RDI to grow high-end white wine grapes in arid climates, especially aromatic varieties, using Riesling as a model.

Impact of winemaking processes on wine polysaccharides, improving by qNMR

Today the knowledge in terms of molecular composition of the colloidal matrix is ​​not enough in order to control its stability, according to the number of winemaking and wine stabilization processes. The physico-chemical processes during the winemaking change the composition and quantity of wine macromolecules. The goal today is to determine which analytical techniques will allow to discriminate these winemaking processes in order to better understand their impact on colloidal matrix stability as well as which molecules are responsible for its instabilities. METHODS: Wines obtained after conventional winemaking were subjected to different fining and chemical stabilization treatments. Different methods were used to investigate the wine macromolecular composition and stability after chemical stabilization, including quantitative and qualitative analyzes of total soluble polysaccharides by extraction under acidified ethanol, and by size exclusion separation as well as qNMR metabolomics. RESULTS: Observation of a slight difference at the quantitative level using classical analysis between the winemaking processes was observed as well as a strong discrimination by qNMR metabolomics.

Response of Albariño to local environmental conditions in Uruguay

Albariño is a white cultivar that has been recently promoted in Uruguay due to its ability to maintain high berry quality even in adverse climate conditions during ripening. This study aims to assess the effect of different topographic conditions on Albariño agronomic behavior and oenological potential.

Sardinia terroir and Cannonau: a zoning approach to discover an ancient tradition

Cannonau variety is historically grown in a large Sardinia area (Jerzu district) and the vineyards are planted both in the plane and in the sloped hills reaching also 650 m of altitude a.s.l. Thus, in order to discover how climate, soil diversity and growing traditions could account for differences in grape and wine quality, this trial was carried out.

Reconfiguring wine prescription : from traditional critics to digital social networks

The integration of digital social networks (DSN) has profoundly transformed communication practices within the wine industry, reorganizing the dynamics of prescription and marketing.

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 the effect of ripening rates on the composition of aroma and phenolic compounds in Cabernet-Sauvignon wines

The study of cultural practices to delay ripening and the characterization of their effect on wine composition is important in the mitigation of accelerated ripening caused by higher temperatures

Above and below: soil moisture and soil temperature interact to alter grapevine water relations

The combined effect of soil moisture and soil temperature on grapevine physiology is gaining interest in the context of global warming.

Effect of nanofiltration on the chemical composition and wine quality

In Enology the conventional processes of filtration for clarification and stabilization are giving place to alternative membrane processes, including nanofiltration (NF). Furthermore, the increased alcohol content in wines recorded in recent years became an important issue for all the main wine producing countries. Among techniques available to the wine industry to reduce the ethanol content, NF is certainly one of the newest. This study is focused on the evaluation of NF influence on wine physical-chemical composition, including mineral content, which in accordance to our best knowledge is a novelty.

Leaf elemental composition in a replicated hybrid grape progeny grown in distinct climates

The elemental composition (the ionome) of grape leaves is an important indicator of nutritional
health, but its genetic architecture has received limited scientific attention. In this study, we
analyzed the leaf ionome of 131 interspecific F1 hybrid progeny from a Vitis rupestris (♀) X Vitis
riparia (♂) cross. The progeny were replicated in New York, South Dakota, Southwest Missouri ad Central Missouri, and the concentration of 20 elements were measured in their leaves at
three different phenological stages during the growing season. In leaves collected at the apical node at anthesis, elemental concentrations correlated in a consistent manner (p < 0.05) across all four geographic locations. In subsequent phenological stages, elemental ratios in the apical-node leaves remained consistent across the South Dakota and New York sites, but not across the Missouri sites. In leaves collected at the basal and middle nodes, correlations varied greatly across all locations.

Influenza delle componenti climatiche e pedologiche sulla variabilità dei contenuti polifenolici in alcuni ambienti vitati della DOCG Sagrantino di Montefalco

Obiettivo del progetto è la valutazione dell’influenza climatica e pedologica dell’areale di Montefalco sul vitigno Sagrantino, ponendo particolare attenzione alla componente polifenolica e antocianica. Sono stati quindi messi a confronto, a partire dal 2001 fino al 2008, sei differenti zone tutte situate all’interno dell’areale DOCG Sagrantino di Montefalco; per ciascun vigneto alla vendemmia sono state effettuate analisi sui parametri analitici e sul contenuto polifenolico e antocianico delle uve. Ognuna delle sei zone è inoltre stata caratterizzata dal punto di vista pedoclimatico, valutando l’influenza del clima e della tipologia di suolo sui parametri analitici presi in considerazione.