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…

Vineyard nutrient budget and sampling protocols

Vineyard nutrient management is crucial for reaching production-specific quality standards, yet timely evaluation of nutrient status remains challenging. The existing sampling protocol of collecting vine tissue (leaves and/or petioles) at bloom or veraison is time-consuming. Additionally, this sampling practice is too late for in-season fertilizer applications (e.g. N is applied well before bloom). Therefore alternative early-season protocols are necessary to predict the vine nutrient demand for the upcoming season. The main goals of this project are to 1) optimize existing tissue sampling protocols; 2) determine the amount of nutrients removed at the end of the growing season.

Non Saccharomyces wine yeasts: emerging trends and challenges in winemaking

In the past, the contribution of non-Saccharomyces yeasts in winemaking has always been considered negative for their limited enological attitude if compared with Saccharomyces cerevisiae. In recent decades there has been a reevaluation of the role of non-Saccharomyces wine yeasts especially when used in combination and in support with S. cerevisiae (mixed fermentation). In this regard, selected non-Saccharomyces yeasts could be profitable used to give distinctive features, to enhance flavor and aroma complexity and to reduce the ethanol content of wines. Further emerging trends in the use of these yeasts are related to their role as bioprotectants and producers of health promoters compounds.

Survey reveals training needs for airblast sprayer operators

In California, little training in sprayer calibration or pesticide drift management is required to apply pesticides. Yet, there is a need to maximize pesticide efficacy and minimize drift. Therefore, our team is developing a training course on airblast application best practices. We distributed a survey to identify current practices and used importance-performance analysis to interpret responses to the importance of spray related topics and satisfaction with previous training.

VALORIZATION OF GRAPE WINE POMACE USING PULSED ELECTRIC FIELDS (PEF) AND SUPERCRITICAL CO₂ (SC CO₂) EXTRACTION

Wine grape pomace quantitatively and qualitatively represents the most important fraction of wine waste. Namely, this by-product makes ~ 20% of the total mass of vinified grapes, and it is characterized with high concentrations of polyphenolic antioxidants, as well as grape seed oil. Hence, valorization of wine pomace, as an alternative to traditionally employed disposal, has drown considerable interest in recent years. Earlier studies were mostly focused on the extraction of phenolics, while mechanisms enhancing the extraction of lipid fraction from grape pomace, as well as their impact on the grape seed oil quality are far less investigated.

Can grapevine tolerance to bunch rot be directly induced by groundcover management?

Botrytis bunch rot occurrence is the most important limitation for the wine industry in humid environments. The effect of grapevine vegetative growth on bunch rot expression results from direct effects (cluster architecture, nitrogen status among others) and indirect ones (via microclimate). Previous studies of our group showed strong differences in bunch rot incidence between floor management treatments: cover crop (CC) vs weed-free strips under the trellis with herbicide (H). We observed that in some circumstances this reduction in bunch rot incidence occurred without major vine growth differences among treatments. The aim of the present study was to test the general hypothesis that other factors unrelated to grapevine vegetative expression could be more relevant to grapevine susceptibility to bunch rot.

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…

Effect of different winemaking techniques and grape variety on chemo-sensory parameters of white wines

AIM: Study the chemical and sensory parameters of fifty commercial white wines elaborated with different techniques (fermented in oak barrel and aged on lees (FB+AL); aged on lees (AL); and without aging (WA)) and different grape varieties (Verdejo, Sauvignon blanc and Godello).

Techniques to study graft union formation in grapevine 

Grapevines are grown grafting in most viticultural regions. Grapevine rootstocks are either hybrids or pure species of different American Vitis spp. (particularly V. berlandieri, V. rupestris and V. riparia), which are primarily used to provide root resistance to the insect pest Phylloxera. In addition to Phylloxera resistance, ideally grapevine rootstocks should be resistant to other soil borne pathogens and adapted to abiotic stress conditions. New rootstocks have the potential to adapt agriculture to climate change without changing the characteristics of the harvested product. However, high grafting success rates are an essential prerequisite.

Effect of quercus alba oak barrels from different forest on the volatile composition of Tempranillo wines

The species and origin used for red wine oak aging determines the physiological composition of the wood and thus the finished wines. In America, oak is grown primarily

Use of mathematical modelling and multivariate statistical process control during alcoholic fermentation of red wine

Cyberphysical systems can be seen in the wine industry in the form of precision oenology. Currently, limitations exist with established infrared chemometric models and first principle mathematical models in that they require a high degree of sample preparation, making it inappropriate for use in-line,

Untargeted LC-HRMS analysis to discover new taste-active compounds in spirits.

​For several years, the chemistry of taste has aroused high interest both from academics and industrials. Plant kingdom is a rich and reliable source of new taste-active compounds. Many sweet, bitter or sour molecules have been identified in various plants [1]. They belong to diverse chemical families and their sensory properties are strongly affected by slight structural modifications. As a consequence, the investigation of natural taste-active products in a given matrix appears as a major challenge for chemists. Such studies are particularly relevant in oenology since they allow a better understanding of wine and spirit taste.