Premium
Classification of brandies by pattern recognition of chemical data
Author(s) -
Miyashita Yoshikatsu,
Ishikawa Masami,
Sasaki ShinIchi
Publication year - 1989
Publication title -
journal of the science of food and agriculture
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.782
H-Index - 142
eISSN - 1097-0010
pISSN - 0022-5142
DOI - 10.1002/jsfa.2740490309
Subject(s) - flavour , pattern recognition (psychology) , artificial intelligence , mathematics , german , computer science , chemistry , food science , geography , archaeology
Classification of German and French brandies according to their geographic origins was achieved by applying pattern recognition techniques to the Reiner and Schreier's composition data of flavour compounds. The Gram‐Schmidt orthogonalisation method and the Fisher ratio selection order were used to select and make new orthogonal features for a useful classification. Two flavour compounds, ethyl octanoate and ethyl butanoate, were found to be most important for the classification of German and French brandies.