Articles
UTILIZATION OF MULTIVARIATE STATISTICS IN EVALUATING MANGO PROCESSING QUALITY
Article number
992_76
Pages
607 – 614
Language
English
Abstract
Mango processing quality is determined by mango raw materials quality.
Quality characteristics of different mango varieties provide significant differences.
Around 10 quality indices, including edible rate, total soluble solids, pH, titratable acid, total sugars, etc. of 16 mango varieties, were analyzed by multivariate statistics methods of factor analysis, principal component analysis (PCA) and systematic cluster analysis.
Results indicated that the cumulative contributions of total variance of the first 5 principal components reached 85.86%. Among them, the first principal compo¬nent included pH and titratable acid, the second included color and fiber, the third included storability and taste, the fourth included total sugar and total soluble solids, the fifth included edible rate and aroma.
That also reflected it was reasonable to evaluate mango processing quality by the 5 principal components.
The result of factor scores indicated the processing quality of Tainong-1, Xiangmang and Machesu were excellent.
Through systematic cluster analysis, 16 mango varieties were divided into 6 groups. Machesu and Guire-284 were classified as group-1 and group-2 respectively, Sri Lanka and Zihua were classified into group-3, Tainong-1, Guire-10 and Guire-82 were classified into group-4, Ivory-22, Xiangmang, etc., were classified into group-5, Hongjinhuang, Qingpi, etc. were classified into group-6. Most varieties in group-2 and group-4 were considered as suitable for processing.
The results showed different mango varieties having similar quality characteristics clustered in the same group.
Quality characteristics of different mango varieties provide significant differences.
Around 10 quality indices, including edible rate, total soluble solids, pH, titratable acid, total sugars, etc. of 16 mango varieties, were analyzed by multivariate statistics methods of factor analysis, principal component analysis (PCA) and systematic cluster analysis.
Results indicated that the cumulative contributions of total variance of the first 5 principal components reached 85.86%. Among them, the first principal compo¬nent included pH and titratable acid, the second included color and fiber, the third included storability and taste, the fourth included total sugar and total soluble solids, the fifth included edible rate and aroma.
That also reflected it was reasonable to evaluate mango processing quality by the 5 principal components.
The result of factor scores indicated the processing quality of Tainong-1, Xiangmang and Machesu were excellent.
Through systematic cluster analysis, 16 mango varieties were divided into 6 groups. Machesu and Guire-284 were classified as group-1 and group-2 respectively, Sri Lanka and Zihua were classified into group-3, Tainong-1, Guire-10 and Guire-82 were classified into group-4, Ivory-22, Xiangmang, etc., were classified into group-5, Hongjinhuang, Qingpi, etc. were classified into group-6. Most varieties in group-2 and group-4 were considered as suitable for processing.
The results showed different mango varieties having similar quality characteristics clustered in the same group.
Publication
Authors
Kejian Zou, Jin Zhang, Li Huang, Dongmei Yan, Jianwen Teng, Baoyao Wei
Keywords
PCA, factor analysis, systematic cluster analysis, quality characteristic, sensory evaluation
Online Articles (76)