Study on the prediction model of basic components on the quality of buckwheat noodles

Abstract The sensory quality of noodles is the key factor in determining consumers' acceptance, and the physicochemical properties can reflect the quality of noodles. In this study, the rheological and thermodynamic properties, noodle quality indexes, and molecular and structural parameters were characterized by adding different levels of buckwheat flour. Pearson correlation analysis was used to evaluate the correlation between physicochemical indexes and basic components of noodles. A comprehensive evaluation model was established by the combination of principal component analysis (PCA) and regression analysis (RA) to evaluate the sensory quality of noodles. The results showed that there was a significant correlation between the physicochemical indexes and the basic components. The two principal components extracted by PCA could explain 89.4% of the total variance of the data. RA showed that the comprehensive evaluation value of the principal component model had a very significant negative correlation with the total score of sensory evaluation ( R 2 = 0.94). In conclusion, this work demonstrated that PCA and RA as an objective protocol had great potential in predicting the sensory quality of noodles.