摘要
卷烟产品中致香成分的种类和含量是区分卷烟风格的主要因素。针对致香成分指标种类繁多的特点,本文提出一种基于多元统计分析的致香成分分析方法。首先采用核主元分析方法将卷烟产品的63种致香成分指标数据映射到二维特征空间,且特征数据的原始信息含量不低于95%;然后引入类间离散度指标分析和评价不同品牌卷烟产品的致香成分相似性。分析结果表明,采用本文方法获得的9个品牌卷烟产品的致香成分相似性分析结果与传统经验评吸结果一致。该方法为传统的卷烟风格分类方法提供了有效的理论支持。
The types and contents of aroma components are the main factors affecting the tobacco style. For a wide range of aroma constituents, this paper presents a new analysis method based on the multivariate statistical analysis. First, the high-dimensional data up to 63 index of tobacco aroma compositions are mapped into two-dimensional feature space by kernel principal component analysis, and the information mount of feature data is not less than 95 %. Then the parameter of inter-class dispersion is introduced to analyze the feature data and evaluate the aroma components similarity of nine brands products. The results showed that the evaluation of nine brands cigarette based on the proposed method was consistent with the Chinese cigarette style feature classification and brand market segmentation. Such method can provide effective theoretical support for the traditional classification method of cigarette-style feature.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2014年第3期377-381,共5页
Computers and Applied Chemistry
基金
云南省应用基础研究计划项目(2011FZ036)
关键词
致香成分
核主元分析
多项式核函数
类间离散度
aroma components
kernel principal component analysis
polynomial kernel function
inter-class dispersion
