Multivariate analysis of quality attributes of sweetpotato flour
| Autor: | Michael Ayodele Idowu, F.O. Henshaw, Taofik Akinyemi Shittu, G. O. Olatunde |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Multivariate statistics
Multivariate analysis Blanching General Chemical Engineering 010401 analytical chemistry 04 agricultural and veterinary sciences Positive correlation 040401 food science 01 natural sciences Industrial and Manufacturing Engineering 0104 chemical sciences Processing methods 0404 agricultural biotechnology Food science Safety Risk Reliability and Quality Food Science Mathematics |
| Zdroj: | Journal of Food Measurement and Characterization. 11:2211-2221 |
| ISSN: | 2193-4134 2193-4126 |
| Popis: | Multivariate methods such as factor analysis (FA), cluster analysis and discriminant function analysis (DFA) were employed in this study to evaluate important parameters for characterization and quality control analysis of sweetpotato flour. Eighty samples of sweetpotato flour were produced from a combination of variety and processing methods. Each sample was evaluated for 35 quality parameters such as chemical, physicochemical and pasting properties. FA resulted in nine factors and identified five pasting viscosities as variables that are more relevant to the variances among the flours. Three clusters were selected which revealed different characteristics among the flour samples. The characteristics provided some information which hitherto were not apparent due to the large data; flours from yellow-fleshed sweetpotato varieties have high paste viscosities while flours from white-fleshed and orange-fleshed varieties have relatively lower paste viscosities; untreated or native flours have high paste viscosities while treatments such as soaking in water, soaking in metabisulphite or blanching caused a lowering of paste viscosities of the flours; sun drying resulted in higher paste viscosities of sweetpotato flours compared to oven drying. DFA showed that 94% of the flours were correctly classified. Two discriminant functions were derived, the use of which improved the classification of the flours. DFA also showed that only four pasting viscosities gave the most efficient combination for classifying the flours, with peak viscosity having the highest positive correlation. This suggests that peak viscosity was the most discriminating property for sweetpotato flour and hence an important index for its quality evaluation. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink