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…

Study of the aromatic oxidation markers of Tempranillo long aged wines

The aromatic quality of wines after a long aging period in bottle is one of key points for oenologists. The objective of this work is to determine the main representative aromatic compounds found in long aged wines from D.O.Ca. Rioja. This study was made by 32 wines from 1971 to 2010 vintages. Sotolon, acetaldehyde, phenylacetaldehyde, 1,1,6-trimethyl-1,2-dihydronaptalene (TDN), β-damascenone, Y-decalactone and Y-dodecalactone were determined as the most important oxidation markers by GC-MS analysis. Moreover, sensory analysis using triangular tests were performed from wines with and without the addition of the mentioned compounds. Four different concentrations of each odorant were added, as individual compounds and as mixtures. The additions were ranged from values close to the reference odour thresholds up to high level concentrations. The most identified aroma was sotolon, which is commonly associated to curry and coffee liqueur aromatic notes. Other oxidative compounds were easily detected by panellists, such as Y-decalactone (peach compote), Y-dodecalactone (ripe fruit). The mixtures of the odorants were most easily detected than the individual compounds. It should be noted that acetaldehyde and phenylacetaldehyde were rarely perceived and distinguished.

Ripening of cv. Cabernet Sauvignon grapes: polysaccharides fractions evolution and phenolic extractability

Polysaccharides and more specifically pectins, make up a significant portion of the cell wall material of the plant cells including the grapes. During the fruit ripening the associated softening is related to the breakdown of the cell wall polysaccharides. During this process, it is expected that polysaccharides that are soluble in red wine will be formed influencing its texture. Anthocyanins are responsible for the wine color and tannins for the astringency, body and bitterness of the wine. In the skins, these compounds are located in the cell vacuoles and the barrier that conditions their extractability is the skin cell wall that may determine the mechanical resistance, the texture and the ease of processing berries. The aim of this work was study the evolution of the polysaccharides and the anthocyanin and tannin extractability during the ripening period in Cabernet Sauvignon grapes, trying to correlate these variables.

A comparative analysis of regions worldwide with Pinot noir

This study examines the growing season climates of selected wine regions worldwide that have significant areas under Pinot noir.

Chardonnay white wine bottled with different oenological tannins: effect on colour traits, volatile composition and sensory attributes during shelf-life

The use of oenological tannins during winemaking has been mostly studied for improving colour traits and stability on red wines. Their effectiveness mainly depends on the tannin composition, grape variety and winemaking approach [1].

Enhanced polyphenol extraction during Pinot Noir and Cabernet Sauvignon wine making

The quality of red wine depends on the composition of polyphenols influencing wine color and taste. The question is, how much we must fear over extraction, especially of seed tannins, under cool climate conditions. The extraction of polyphenols from grape skins and grape seeds were investigated for the grape varieties Cabernet Sauvignon and Pinot noir

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…

Measurement of trans-membrane and trans-tissue voltages in the Shiraz berry mesocarp

In mid to late ripening, sugar and potassium (K+) accumulation into the berry slows and is eventually completed1. K+ is the most abundant cation in the berry, undertaking important physiological roles.

Terroirs and legal protection

Le concept AOC permet, par une délimitation précise, la mise en valeur de terroirs particulièrement adaptés à la viticulture. Seuls les terroirs ainsi identifiés peuvent produire des vins portant le nom de l’AOC. Le nom de cette AOC ne peut être utilisé que pour des vins issus de terroirs compris dans l’aire d’appellation, sous peine de sanctions pénales. La délimitation ainsi opérée participe à la protection du nom de l’AOC. A l’inverse, le terroir délimité n’est pas protégé.

Vine field monitoring using high resolution remote sensing images: segmentation and characterization of rows of vines

A new framework for the segmentation and characterization of row crops on remote sensing images has been developed and validated for vineyard monitoring. This framework operates on any high-resolution remote sensing images since it is mainly based on geometric information. It aims at obtaining maps describing the variation of a vegetation index such as NDVI along each row of a parcel.

Determination of metallic elements in Chilean wines by atomic absorption spectroscopy and inductively coupled plasma–mass spectrometry

The chemical composition of wines depends on series of variables such as the type of grape, edaphoclimatic conditions, and viticulture and winemaking practices employed during production. Metallic elements play a significant role during winemaking (e.g. as catalysts of oxidation reactions) and have been previously employed for the classification of wines according to provenance. In this work, we focused on the analysis of metallic elements (K, Na, Ca, Zn, Cu, Fe, Mg, Mn, Ni, Cr, Al, Pb, Cd, Hg, Se, Co, Sn and As) in 145 Chilean wine samples (102 reds and 43 white wines), of seven grape varieties, and five of the major wine producing regions in Chile.

Methodology for soil study and zoning

La caractérisation des sols en vue d’une étude de terroirs viticoles peut être réalisée à différents niveaux de complexité, suivant le nombre de variables pris en compte et suivant le fait que celles-ci sont spatialisées ou non