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…

Effect of different winemaking practices on chemical composition, aroma profile and sensory perception of ribolla gialla sparkling wines

This study aims at evaluating the effects of different refermentation methods (Martinotti/Charmat vs. Classic) on the chemical composition, aroma profile and sensory characteristics of Ribolla Gialla sparkling wines; furthermore, certain winemaking practices (skin contact and use of pectolytic enzymes) were investigated considering the extraction of varietal aromas and aroma precursors. METHODS: Sparkling wines were produced at pilot-plant scale. Concerning refermentation methods, traditional Martinotti (MB – 30 days length), extended Martinotti (ML) with 4 months of aging on lees and Classic method (CL) with 11 months of aging on lees were compared; in a second trial, skin contact (MM), enzyme addition on must also subjected to maceration (ME), and enzyme addition on base wine (VE) were evaluated. All experimental trials were performed in triplicate. Basic chemical composition, varietal (terpenes and C13-norisoprenoids in free and bound form) and non-varietal aroma compounds were evaluated by LLE-GCMS analysis; finally, sensory analysis was also performed, by descriptive testing.

Genome wide association mapping of phenology related traits in Vitis vinifera L

Climate change, with rise in temperatures, is leading to an advance in the dates of phenological stages, with a loss in quality of the grape final product. Therefore, the understanding of the genetic determinants driving the phenological stages of flowering, veraison and the interval between them, represents a target for the development of grapevine’s cultivar adapted to the changing environment.
Here we conducted a GWA study to identify SNPs significantly associated to flowering time, veraison time and to the interval among them. A germplasm collection (CREA-VE in Susegana, Treviso, Italy) including 649 grapevine’s cultivar representing 365 unique genotypes was considered.

The exploitation of Croatian grapevine genetic resources for the breeding of new resistant cultivars 

Croatian viticulture is mainly based on native grapevine varieties susceptible to various diseases and pests, which leads to unsustainable use of large amounts of pesticides. The sustainable development of viticulture in the future will only be possible by increasing the resistance of the grapevine through the development of new resistant varieties. Breeding programs have been launched in the leading wine-growing countries to develop resistant varieties possessing high-quality levels. Native cultivars from Croatia are not included in the breeding programs of other countries.

Permanent cover cropping with reduced tillage increased resiliency of wine grape vineyards to climate change

Majority of California’s vineyards rely on supplemental irrigation to overcome abiotic stressors. In the context of climate change, increases in growing season temperatures and crop evapotranspiration pose a risk to adaptation of viticulture to climate change. Vineyard cover crops may mitigate soil erosion and preserve water resources; but there is a lack of information on how they contribute to vineyard resiliency under tillage systems. The aim of this study was to identify the optimum combination of cover crop sand tillage without adversely affecting productivity while preserving plant water status. Two experiments in two contrasting climatic regions were conducted with two cover crops, including a permanent short stature grass (P. bulbosa hybrid), barley (Hordeum spp), and resident vegetation under till vs. no-till systems in a Ruby Cabernet (V. vinifera spp.) (Fresno) and a Cabernet Sauvingon (Napa) vineyard. Results indicated that permanent grass under no-till preserved plant available water until E-L stage 17. Consequently, net carbon assimilation of the permanent grass under no-till system was enhanced compared to those with barley and resident vegetation. On the other hand, the barley under no-till system reduced grapevine net carbon assimilation during berry ripening that led to lower content of nonstructural carbohydrates in shoots at dormancy. Components of yield and berry composition including flavonoid profile at either site were not adversely affected by factors studied. Switching to a permanent cover crop under a no-till system also provided a 9% and 3% benefit in cultural practices costs in Fresno and Napa, respectively. The results of this work provides fundamental information to growers in preserving resiliency of vineyard systems in hot and warm climate regions under context of climate change.

Effect of the plant sink/source balance on the chemical content of red table grapes (Vitis vinifera L.).

PPhloem transport of assimilates provides the materials needed for the growth and development of reproductive structures, storage and developing organs, and has long been recognized as a major determinant in crop yield.

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…

Market analysis of Chilean Pinot noir, Carménère, and Cabernet-Sauvignon wines: A comparative study of chemical parameters across low, medium, and high price segments

Wine quality is a complex concept determined by multiple factors, including vineyard management, winemaking operations, and the sensory perception of key attributes.

Multispectral data from Sentinel-2 as a tool for monitoring late frost events on vineyards

Aim: Climate change is altering some aspects of winegrape production with an advancement of phenological stages which may endanger viticultural areas in the event of a late frost. This study aims to evaluate the potential of satellite-based remote sensing to assess the damage and the recovery time after late frost events.

Influence of soil characteristics on vine growth, plant nutrient levels and juice properties: a multi-year analysis

Soil physical and chemical properties affect vine nutrition, as indicated by leaf and petiole nutrient content, in a way that may directly impact wine properties.

Are all red wines equals regarding their vulnerability to Brettanomyces bruxellensis ?

Odours deemed harmful by the consumer and described as “stable”, “horse sweat” or “burnt plastic” can be found in wines. The responsible molecules are volatile phenols, produced by a spoilage yeast: brettanomyces bruxellensis. This species is particularly well adapted to the wine environment and can resists many stresses such as a high alcohol level, a low ph or high levels of SO2, more or less efficiently depending on the strain considered.

Diffuse light due to wildfire smoke enhances gas exchange of shaded leaves

The risk of wildfires is increasing as the frequency and severity of drought and heat waves continue to rise. Wildfires are associated with the combustion of plant materials and emit smoke. In the atmosphere, smoke may spread readily across large areas. Smoke is composed of solid and liquid phase particulates and gases and has been identified as a causal agent of “smoke taint” in wine. On a smoky day, the intensity of direct light decreases because these particulates scatter sunlight. Even though this effect is frequently assumed to decrease plant photosynthesis, this assumption ignores the potential changes in diffuse light and may be based on scant evidence.