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…

Oenological potential of wines and agronomical characterisation of grapes from five white resistant Italian varieties at Serra Gaúcha, Southern Brazil

Rio grande do sul is the main grape producing state in Brazil, with the largest wine-growing area, responsible by 90% of the national production of wines and grape juices. Serra Gaúcha is the main vitivinicultural region, where around 15% of the area is destined to produce wines from vitis vinifera L. grapes. This region presents high rainfall during the grape maturation cycle, a factor that leads to great risk of attacks by fungal pathogens. the use of resistant varieties can reduce the cost and quantity of spraying, improving wine quality, focusing on a sustainable vitiviniculture.

New technologies to characterize spatial variability in viticulture

Measurements of parameters spatialy positionned, with on line sensors mounted on classical machinery or airborne imagery is no more a problem in viticulture. In a short time, high resolution data dedicated to the assessment of the vine characteristics, the soil, the harvest, etc. will become a reality.

Organic mulches slightly influence wine phenolic composition and sensorial properties

Grapevines have traditionally been grown in semi-arid areas, but viticulture is now compromised by climate change. Therefore, it is necessary to implement environmentally friendly viticulture practices to adapt grapevines to current climatic conditions. In this context, organic mulches offer many benefits, such as reduced soil erosion and increased organic matter, soil water content and crop productivity. However, these practices must not compromise grape and wine quality. Therefore, the objective of this study was to evaluate the effect on wine physicochemical and phenolic composition and sensorial properties of different soil management practices on the vine row. Over four years, five soil treatments were examined in two different vineyards.

First characterization of thiol precursors in colombard and gros manseng: comparison of two cultivation practices

AIM: Organic production of wine in the past years has known an important augmentation. This type of cultivation practice switches synthetic phytosanitary product for copper-based protection as fungicide.

«Observatoire Mourvèdre»: statistical modelling of quality for Cv. Mourvèdre

Vine cultivar Mourvèdre is present all around the Mediterranean area and is interesting for its tannins and the specificity of its aromas.

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…

Monitoring of Sangiovese red wine chemical and sensory parameters along one-year aging in different tank materials and glass bottle

The aim of this study was to test how different tank materials could affect the chemical composition and the sensory profile of a red wine during an entire year of aging. For this scope, a single varietal Sangiovese wine was aged, after completing its malolactic fermentation, by using tanks made by different materials. Six thesis were involved in the aging experiment, in particular: stainless

Is it relevant to consider remote sensing information for targeted plant monitoring?

An experiment was carried out to test the relevance of using satellite images (NDVI) to define locations of plant monitoring systems.

Application to the wine sector of European Convention on the landscapes

The landscape is defined by the European convention of the landscape (Florence, October 20, 2000) like part of the territory as perceived by the populations, whose character results from the action of natural and/or human factors and their interrelationships. This convention is based on the contribution cultural, ecological, environmental, social of the landscapes and aims at a reinforcement of the tools of protection and valorization in particular in the agricultural policies, of regional planning and town planning. Moreover, it encourages a step of identification and qualification of the landscapes and underlines the need for developing the sensitizing and the training of the actors concerned.

Effects of future climate change on grape quality: a case study for the Aglianico grape in Campania region, Italy

Water deficits limit yields and this is one of the negative aspects of climate change. However, this applies particularly when emphasis is on biomass production (e.g. for crops like maize, wheat, etc.)

Elicitors used as a tool to increase stilbenes in grapes and wines

The economic importance of grapevine as a crop plant makes Vitis vinífera a good model system to study the improvement of the nutraceutical properties of food products (Vezulli et al. 2007). Stilbenes in general, and trans-resveratrol in particular, have been reported to be responsible for various beneficial effects. Resveratrol´s biological properties include antibacteria and antifungal effects, as well as cardioprotective, neuroprotective and anticâncer actions (Guerrero et al. 2010 ). Stilbenes can be induced by biotic and abiotic elicitors since they are phytoalexins (Bavaresco et al. 2001).