葡萄酒的评价模型  

The Evaluation Model of Wine

作　　者：彭定忠[1] 彭翔[2] 孙敏敏[2] 肖蓝湘[3] 

机构地区：[1]湖南理工学院数学学院,湖南岳阳414006 [2]湖南理工学院信息与通信工程学院,湖南岳阳414006 [3]湖南理工学院机械工程学院,湖南岳阳414006

出　　处：《湖南理工学院学报（自然科学版）》2012年第4期10-16,共7页Journal of Hunan Institute of Science and Technology(Natural Sciences)

摘　　要：对两组评酒员的平均打分进行双样本t检验及kendall协调系数W检验,最终确定第一组评酒员更可信;然后利用基于主成分的聚类分析和评酒员对酒样平均打分,实现酿酒葡萄分类定级;采用基于主成分的多分类Logistic回归建立酿酒葡萄与葡萄酒的理化指标之间的数学模型;通过制定葡萄酒分类标准,利用基于主成分的多分类Logistic回归建立酿酒葡萄和葡萄酒的理化指标与葡萄酒等级的数学模型.Firstly, through two-sample t test and Kendall coordination coefficient W test about two groups of taster＇s average score, the first group of taster is considered to be more credible, then the classification and rank of wine grape is realized based on the principal component＇s cluster analysis and taster＇s average score; afterwards, the mathematical model about physicochemical index between wine grape and wine is established based on the principal component＇s multinomial Logistic regression, finally through setting criterion for wine classification, the mathematical model about the physicochemical index of the wine grape and wine and the level of wine is built based on the principal component＇s multinomial Logistic regression.

关 键 词：双样本t检验 kendall协调系数W检验 主成分分析 聚类分析 多分类Logistics回归 

分 类 号：O213[理学—概率论与数理统计]