Scholarlink

<p>Predicting wine quality is increasingly challenging due to global warming and the intensification of extreme weather events. Traditional climatic indices often fail to capture the impact of these anomalies. This paper addresses this gap by describing a system that not only monitors climatic data but explicitly incorporates detected weather anomalies into a predictive model for Brunello di Montalcino. The system allows visualization and highlighting of anomalous observations within environmental time-series data from vineyards. We exploit these detected anomalies and historical climatic data to determine an optimal combination of indices for predicting future wine quality. Using multiple linear regression models correlated against expert quality ratings, we find that a combination of the Huglin Index, mean humidity, and mean wind velocity achieves a strong predictive result. We demonstrate that integrating anomaly detection data as a weighted feature in the regression model improves the final prediction, achieving a coefficient of determination (<InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\varvec{R^2}\)</EquationSource> </InlineEquation>) of 0.546.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of wine quality by monitoring climatic data and weather anomalies

  • 期刊论文

  • 发表日期：

    2026年3月24日

查看全文
  • Federico Becattini,
  • Andrea Ferracani,
  • Giuseppe Becchi,
  • Alberto Del Bimbo

摘要

Predicting wine quality is increasingly challenging due to global warming and the intensification of extreme weather events. Traditional climatic indices often fail to capture the impact of these anomalies. This paper addresses this gap by describing a system that not only monitors climatic data but explicitly incorporates detected weather anomalies into a predictive model for Brunello di Montalcino. The system allows visualization and highlighting of anomalous observations within environmental time-series data from vineyards. We exploit these detected anomalies and historical climatic data to determine an optimal combination of indices for predicting future wine quality. Using multiple linear regression models correlated against expert quality ratings, we find that a combination of the Huglin Index, mean humidity, and mean wind velocity achieves a strong predictive result. We demonstrate that integrating anomaly detection data as a weighted feature in the regression model improves the final prediction, achieving a coefficient of determination ( \(\varvec{R^2}\) ) of 0.546.