Abstract:

Fuzzy mathematics and cluster analysis were used for comprehensive quality assessment of fourteen table grape
varieties sampled from a commercial vineyard in Weinan, Shannxi province in 2013. The results showed that there were
larger coefficients of variation in berry shape, yield, total anthocyanin content, berry skin thickness, berry bloom thickness,
and berry color characteristics, whereas sensory properties and related indicators of intrinsic quality were relatively
lower. The comprehensive assessment by fuzzy mathematics analysis showed that “Red Globe”, “Summer Black”, “Ruby
Seedless”, “Wuhengzaohong” ,“Jintian 0601” and “Wink” had superior quality. “Qinlongdasui” and “Flame Seedless”
showed the lowest quality. By using Q-type cluster analysis, “Wuhengzaohong”, “Red Globe”, “Ruby Seedless” and “Wink”
with superior quality were clustered into one group, “Muscat Hamburg”, “Jumeigui” and “Hong Ru” with average quality
were assigned into another group, and “Flame Seedless” with the worst quality was clustered into a separate category. The
combination of fuzzy mathematics and cluster analysis could provide a comprehensive and objective evaluation system for
determining the quality of grape varieties.

Key words: grape varieties, comprehensive quality, fuzzy mathematics, cluster analysis