当前位置: X-MOL 学术Chemometr. Intell. Lab. Systems › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Kurtosis-based projection pursuit analysis to extract information from sensory attributes of cachaça
Chemometrics and Intelligent Laboratory Systems ( IF 3.7 ) Pub Date : 2020-08-01 , DOI: 10.1016/j.chemolab.2020.104075
Eduarda Barbieri da Silva Cruz , Michel Rocha Baqueta , Raphael Menechini Neto , Manuel Salvador Vicente Plata Oviedo , André da Silva Castilhos de Melo , Paulo Henrique Março , Patrícia Valderrama , Stephen Driscoll , Peter D. Wentzell

Abstract Kurtosis-based projection pursuit (kPPA) coupled with a variable selection was employed to evaluate twenty cachaca samples concerning the sensory attributes. Cachacas aged in wood barrels from Minas Gerais state, and artisanal flavored cachacas from Morretes, a historical city in the Parana state were evaluated. The selection of sensory attributes by simple variable selection improved the pattern recognition by searching for combinations of the sensory attributes that have a minimum kurtosis value when kPPA is applied. The variable selection highlights the sensory attributes related to the taste (bitter, sweet, and astringency), and related to the smell (caramel, alcoholic, and wood). This approach was able to show that flavoring can help with cachaca costs reduction related to the wood barrels and produce some sensory attributes with notes similar to those aging in wood barrels. A kurtosis-based projection pursuit is a powerful tool for extract information from the sensory analysis.

中文翻译:

基于峰度的投影追踪分析从卡查萨的感官属性中提取信息

摘要 基于峰度的投影追踪 (kPPA) 与变量选择相结合,用于评估 20 份关于感官属性的 cachaca 样本。对在米纳斯吉拉斯州的木桶中陈酿的 Cachacas 和来自巴拉那州历史名城莫雷特斯的手工风味 cachacas 进行了评估。通过简单的变量选择来选择感官属性,通过搜索应用 kPPA 时具有最小峰度值的感官属性组合来改进模式识别。变量选择突出了与味道(苦、甜和涩)相关和与气味(焦糖、酒精和木头)相关的感官属性。这种方法能够表明,调味可以帮助降低与木桶相关的可卡卡成本,并产生一些与木桶陈酿类似的感官特性。基于峰态的投影追踪是从感官分析中提取信息的强大工具。
更新日期:2020-08-01
down
wechat
bug