摘要
根据酿酒葡萄与葡萄酒的理化指标,按"逐步降维,保留信息"的思想建立模型,把原来多个变量划为几个综合指标,并对这些综合指标进行研究,以简化问题的求解过程;因酿酒葡萄和葡萄酒质量间的内在联系及变化规律具有随机性、理化指标信息过大,先进行变量替换,分别选取15个、8个综合变量替代原变量,再对酝酿葡萄与酝酿葡萄酒主成分分析,最后进行典型相关分析。
According to physical and chemical indicators of wine grape and wine, the model is set up based on the idea of dimension reduction step by step and retaining the information to classify original multi-variable to several comprehensive indicators and to study these comprehensive indicators so as to simplify the solving process of the problems. Because the intrinsic connection and changing law between wine grapes and wine quality have randomness and because their physical and chemical indicators are too large, this paper firstly makes variable substitution by selecting 15 comprehensive variables and 8 comprehensive variables respectively to replace variables, then conducts primary component analysis on wine grapes and wine and correlation analysis. finally original carries out typical
出处
《重庆工商大学学报（自然科学版）》
2014年第1期93-97,共5页
Journal of Chongqing Technology and Business University:Natural Science Edition
关键词
降维
主成分分析
典型相关分析
变量替换
dimension reduction
principal component analysis
variable substitution
