标题:
基于数据挖掘技术的葡萄酒评价体系研究The Study on Evaluation System of Wine Based on Data Mining
作者:
王思哲, 王志刚, 何勇
关键字:
葡萄酒评价, 数据挖掘技术, 方差分析, 聚类分析, 回归分析Evaluation System of the Wine, Data Mining Technology, One-Way ANOVA, Cluster Analysis, Regression Analysis
期刊名称:
《Advances in Applied Mathematics》, Vol.4 No.4, 2015-11-30
摘要:
本文以2012年高教社杯全国大学生数学建模竞赛A题为例，利用数据挖掘技术建立葡萄酒评价体系。葡萄酒质量一般是通过聘请有资质的品酒员进行品尝评分，由于品酒员主观因素导致对酒样品的评分差异悬殊，我们通过方差分析对品酒员评分进行可信性研究；通过提取酿酒葡萄多个理化指标的公共因子，对因子得分和葡萄酒评分进行聚类，将酿酒葡萄进行分级研究；通过对葡萄酒理化指标和酿酒葡萄理化的数据进行相关性分析，利用逐步回归分析模型建立它们之间的依赖关系；利用葡萄酒芳香物质与葡萄酒评分之间的回归模型，找出决定葡萄酒质量的关键理化指标，最后指出现行葡萄酒评分体系的不足。
Based on Question A of Mathematical Contest in Modeling for college students in 2012, the empha-sis in this paper is mainly on the establishment of evaluation system of wine based on data mining technology. The wine quality is determined by the score of the wine tasting. We analyze the credi-bility of the liquor score by one-way ANOVA. We classify the wine grape by extracting common factors of some physical and chemical indicators from the wine grape, and by clustering the factor score and wine score. The stepwise regression model is established through the correlation be-tween the physical and chemical indicators and the physical and chemical indicators of wine grapes. By the regression model between the aroma substances and the score of the wine, the key physical and chemical indicators of wine quality will be found. In the end, some shortcomings of current rating system of wine will be pointed out.