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…

Marketing terroir wines

The markets for quality wine are becoming more competitive as newer producers emerge and traditional producers improve their quality. The concept of terroir is one way to differenzi­ate wines in a competitive market and to enhance producer income.

Kinetic study of browning caused by laccase activity using different substrates

To our knowledge all the studies about laccase kinetics and its inhibition have been performed with substrates and conditions very different from those of real grape juice. Moreover, none of these researches really measure enzymatic browning, since they have not taken into account what happens after the oxidation of o-diphenols in o-diquinones and their subsequent polymerization to form melanins1. For that reason, the aim of this research was to develop a new model to measure the kinetics of browning caused by Botrytis cinerea laccase under conditions much closer to those of grape juice and using the substrates naturally present in it.

Viticultural Climatic Zoning and Digital Mapping of Rio Grande do Sul – Brazil, using Indices of the Géoviticulture MCC System

The State Rio Grande do Sul is the main producer of Brazilian fine wines, with four viticultural regions. The objective is the characterization of the viticultural climatic potential of the State (total surface of 281.749 km2). The methodology use the Géoviticulture Multicriteria Climatic Classification System (Géoviticulture MCC System), based on three climatic indices – Dryness Index (DI), Heliotermal Index (HI) and Cool Night Index (CI).

Stomatal behaviour of three minority grapevine varieties grown in the La Mancha region (Spain)

[lwp_divi_breadcrumbs home_text="IVES" use_before_icon="on" before_icon="||divi||400" module_id="publication-ariane" _builder_version="4.19.4" _module_preset="default" module_text_align="center" module_font_size="16px" text_orientation="center"...

Isohydric and anisohydric behavior of 18 wine grape varieties grown in an arid climate

The interest in understanding the water balance of terrestrial plants under drought has led to the creation of the isohydric/anisohydric terminology. The classification was related to an implication-driven framework, where isohydric plants maintain a constant and high leaf water potential through an early and intense closure of their stomata, hence risking carbon starvation. In contrast, anisohydric plants drop their leaf water potential to low values as soil drought is establishing due to insensitive stomata and thus risk mortality through hydraulic failure, albeit maximizing carbon intake. When applied to grapevines, this framework has been elusive, yielding discrepancies in the classification of different wine grape varieties around the world.

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…

Correlation between skin cell wall composition and phenolic extractability in Cabernet sauvignon wines

The phenolic component of red wine is responsible for important elements of flavor and mouthfeel, and thus quality of the finished wine. Additionally, many of these phenolics have been associated with health benefits such as reduction of the risk of developing cardiovascular disease, cancer, osteoporosis and preventing Alzheimer’s disease. While the origins, concentrations, and chemistries of the phenolics in a finished red wine are well known, the fundamental mechanisms and kinetics of extraction of these phenolics from grape skins and seeds during red wine fermentation are poorly understood. This lack of knowledge regarding the extraction mechanisms of phenolics during red wine fermentation makes informed manipulations of the finished wine’s phenolic composition difficult.

Impact of aspects of the polysaccharide structure of mannoproteins on their interactions with Enological Tannins

Mannoproteins (MPs) with different structure of their polysaccharide part (branching, substitutions, …) were used to better understand the impact of characteristics of the usual structure of MPs when interacting with Grape Seed Tannins (ST). 

Cover crops influence on soil N availability and grapevine N status, and its relationship with biogenic

The type of soil management, tillage versus cover crops, can modify the soil microbial activity, which causes the mineralization of organic N to NO3–N and, therefore, may change the soil NO3–N availability in vineyard. The soil NO3–N availability could influence the grapevine nutritional status and the grape amino acid composition. Amino acids are precursors of biogenic amines, compounds mainly formed during the malolactic fermentation. Biogenic amines have negative effects on consumer health and on the wine organoleptic quality. The objective was to study if the effect of conventional tillage and two different cover crops (leguminous versus gramineous) on grapevine N status, could relate to the wine biogenic amines composition.

Enological evaluation of the attitude of the grapevine fumin to give varietal wines

Initiatives have been ongoing in recent years to safeguard biodiversity in the oenological sector via a process of enhancement of ancient varieties, under a pressure of a market strongly oriented towards production deriving from native vines of specific geographical zones. In that sense, Aosta Valley
(Italy) has raised the need to preserve and characterize its minority vine varieties which have the potentiality to give varietal wines. Fumin represents the 7% of the production of the region with 16 hectares of vineyards and 753 hectolitres of derived wine. Due to its large phenolic potential, strong astringency and deep colour, it has long been, and is still today, assembled or blended with other varieties as occurs, for example, for the Torrette.

Spectral characterisation of fungal diseases on Vitis vinifera leaves

Aims: The aims of this study were to (1) detect alterations in the reflectance spectra of vines with fungal diseases, (2) map these alterations, and (3) determine the best wavelengths which may be used as early indicators of fungal diseases in vines.