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…

Insights from selected ion flow tube mass spectrometry (SIFT-MS) and chemometrics applied to the quick discrimination of grapevine varieties

Selected Ion Flow Tube Mass Spectrometry (SIFT-MS) is an innovative analytical method based on soft chemical ionization to analyze thecomposition in volatile compounds of a gas phase

Stem growth disorder and xylem anatomy modifications during esca pathogenesis in grapevines

Esca is a grapevine vascular disease with detrimental consequences on vineyard yield and longevity. Recently, esca leaf symptom development has been shown to result in the occlusion of xylem vessels by tyloses in leaves and stems, leading to hydraulic failure. However, little is known regarding the response of xylem anatomy and stem growth to esca in different varieties . Here we studied the impact of esca leaf symptom development on grapevine physiology, stem growth, and xylem anatomy in two widespread cultivars, Cabernet sauvignon and Sauvignon blanc.

Yield formation and grape composition: more than meets the eye 

Fruit quality in grapes is not well defined but is often depicted as correlating inversely with crop yield. Both fruit yield and composition, however, are made from distinct components that interact in complex ways. Reproductive growth of grapevines extends over two growing seasons. Inflorescences initiated in buds during the previous year differentiate flowers and set and develop berries during the harvest year.

Identification of loci associated with specialised metabolites in Vitis vinifera

Secondary (or specialised) metabolites such as terpenes and phenolic compounds are produced by plants for various roles which include defence against pathogens and herbivores, protection against abiotic stress, and plant signalling. Additionally, these metabolites influence grapevine quality traits such as colour, aroma, taste, and nutritional value. However, the biosynthesis of these metabolites is often complex and controlled by multiple genes which in grapevine are predominantly uncharacterised.

Aroma accumulation trends during berry development and selection of grape aroma candidate genes suitable for functional characterization

Grape flavour management in the vineyard requires knowledge of the derivation of individual flavour and aroma characteristics and the effects that different concentrations and interactions between these compounds have on flavour potential.

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…

VitiProtect–Development and testing of a downy mildew AI forecasting model for Swiss viticulture

Downy mildew (Plasmopara viticola) is a fungal pathogen that causes a destructive disease in grapevines (Vitis vinifera).

Valpolicella chemical pattern of aroma ‘terroir’ evolution during aging

Valpolicella is an Italian region famous for the production of high quality red wines. Wines produced in its different sub-regions are believed to be aromatically different, as confirmed by recent studies in our laboratory. Aging is a very common practice in Valpolicella and it is required by the appellation regulation for periods up to four years. The aim of this study was to investigate the evolution, during aging, of volatile chemical composition of Valpolicella wines obtained from grapes harvested in different sub-regions during different vintages.

ALCOHOLIC FERMENTATION AND COLOR OF ROSÉ WINES: INVESTIGATIONS ON THE MECHANISMS RESPONSIBLE FOR SUCH DIVERSITY

Color is one of the key elements for the marketing of rosé wines due to their packaging in transparent bottles. Their broad color range is due to the presence of pigments belonging to phenolic compounds extracted from grapes or formed during the wine-making process. However, the mechanisms responsible for such diversity are poorly understood. The few investigations performed on rosé wines showed that their phenolic composition is highly variable, close to that of red wines for the darkest rosés but very different for light ones [1]. Moreover, large variations in the extent of color loss taking place during fermentation have been reported but the mechanisms involved and causes of such variability are unknown.

Shoot positioning: effect on physiological, vegetative and reproductive parameters

[English version below]

On a étudié durant deux saisons de croissance (2002/2003 et 2003/2004) l’effet de l’orientation vertical des rameaux sur les paramètres physiologiques, végétatifs et reproductifs dans la région de Stellenbosch dans un vignoble du cépage Merlot sur 99 R conduite à espalier et taillé a cordon coursonné. Les vignes étaient espacées 2.7 x 1.5 m. L’irrigation a été appliquée quand la baie avait la dimension d’un pois et a la véraison.

The interplay between grape ripening and weather anomalies – A modeling exercise

Current climate change is increasing inter- and intra-annual variability in atmospheric conditions leading to grapevine phenological shifts as well altered grape ripening and composition at ripeness. This study aims to (i) detect weather anomalies within a long-term time series, (ii) model grape ripening revealing altered traits in time to target specific ripeness thresholds for four Vitis vinifera cultivars, and (iii) establish empirical relationships between ripening and weather anomalies with forecasting purposes. The Day of the Year (DOY) to reach specific grape ripeness targets was determined from time series of sugar concentrations, total acidity and pH collected from a private company in the period 2009-2021 in North-Eastern Italy. Non-linear models for the DOY to reach the specified ripeness thresholds were assessed for model efficiency (EF) and error of prediction (RMSE) in four grapevine cultivars (Merlot, Cabernet Sauvignon, Glera and Garganega). For each vintage and cultivar, advances or delays in DOY to target specified ripeness thresholds were assessed with respect to the average ripening dynamics. Long-term meteorological series monitored at ground weather station by means of hourly air temperature and rainfall data were analyzed. Climate statistics were obtained and for each time period (month, bimester, quarter and year) weather anomalies were identified. A linear regression analysis was performed to assess a possible correlation that may exist between ripening and weather anomalies. For each cultivar, ripeness advances or delays expressed in number of days to target the specific ripening threshold were assessed in relation to registered weather anomalies and the specific reference time period in the vintage. Precipitation of the warmest month and spring quarter are key to understanding the effect of climate change on sugar ripeness. Minimum temperatures of May-June bimester and maximum temperatures of spring quarter best correlate with altered total acidity evolution and pH increment during the ripening process, respectively.