Predictability of price of tea from sensory assessments and biochemical information using data-mining techniques

Paul, Sanjoy K. (2008) Predictability of price of tea from sensory assessments and biochemical information using data-mining techniques. Journal of the Science of Food and Agriculture, 88 8: 1354-1362. doi:10.1002/jsfa.3223


Author Paul, Sanjoy K.
Title Predictability of price of tea from sensory assessments and biochemical information using data-mining techniques
Journal name Journal of the Science of Food and Agriculture   Check publisher's open access policy
ISSN 0022-5142
1097-0010
Publication date 2008-06
Sub-type Article (original research)
DOI 10.1002/jsfa.3223
Volume 88
Issue 8
Start page 1354
End page 1362
Total pages 9
Editor David S. Reid
Place of publication West Sussex, United Kingdom
Publisher John Wiley & Sons
Language eng
Subject 11 Medical and Health Sciences
Formatted abstract
BACKGROUND: The valuation of tea depends on the sensory assessments made by the Brokers and Buyers (Tea Tasters) to a large extent, though the market conditions and the requirements of a particular Buyer play an important role in determining the basic prices of teas. Again, there are several biochemical quality parameters in tea on which the quality of a particular tea depends. It is not straightforward to establish the reflection of biochemical quality characteristics in tea on the Taster’s sensory assessments and price because of the complex dynamics within chemical properties and the inherent subjectivity of quality evaluation through the Taster’s scores. It is, however, important to judge the market valuation of teas from quality assessments and biochemical properties. This paper describes the advantages of using statistical data-mining techniques to explore the association of biochemical quality parameters in teas with the Taster’s sensory assessments, and the application of a nonparametric statistical technique, multivariate adaptive regression splines (MARSplines), to establish the predictability of the realised prices of teas from sensory assessments.
RESULTS: The price of tea is significantly associated with various quality attributes and some of the biochemical parameters. The MARSplines technique successfully demonstrated the predictability of price through Tasters’ sensory assessments and also raised the issue of inherent subjectivity of the Tasters’ assessments. CONCLUSION: It is important to explore appropriate statistical techniques to assess the subjectivity in Tasters’ assessments, and a better-designed study needs to be conducted to understand the complex biochemical reflections on the price of tea.
Keyword Biochemical quality parameters
Sensory assessments
Tea price
Data mining
Regression splines
Q-Index Code C1
Q-Index Status Provisional Code
Institutional Status Non-UQ

Document type: Journal Article
Sub-type: Article (original research)
Collections: Excellence in Research Australia (ERA) - Collection
School of Public Health Publications
 
Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 1 times in Thomson Reuters Web of Science Article | Citations
Scopus Citation Count Cited 2 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Tue, 05 Jan 2010, 09:07:04 EST by Ms Susana Macanawai on behalf of Faculty Of Health Sciences