Abstract: 【Objective】 This study adopted different methodologies to identify the key texture characteristics and physicochemical indexes affecting the total sensory quality of pineapple, and to establish a new comprehensive evaluation method  for the precise testing of fruit texture and taste quality. 【Method】 In this study, seven different varieties of pineapple (Ananas comosus (Linn.) Merr.) were selected for the measurement of sensory attributes, texture characteristics, and physicochemical compositions. Based on the variance and  correlation analysis, the key texture properties and physicochemical indexes that affecting sensory quality were identified. Further, principal component regression analysis was performed, with key texture characteristics and physicochemical indexes selected as independent variables, and the total score of sensory evaluation as dependent variables, to obtain a statistically significant prediction model for the comprehensive evaluation of pineapple quality.【Result】There are significant differences in some texture properties and physicochemical indexes among different varieties of pineapple. Such as hardness, chewability, maximum shear force, sugar-acid ratio, soluble protein, vitamin C and soluble pectin, the coefficient of variation among varieties was greater than 25 %, while the difference in elasticity and cohesiveness is not significant among varieties. The overall satisfaction score of different pineapple varieties from the highest to lowest is Tainong 17 > Tainong 16 > Tainong 4 > MD-2 > Tainong 11 >Smooth Cayenne > Comte de Paris. Tainong 17 showed the best quality of fruit texture and taste, its total soluble solid content is 16.23 %, sugar to acid ratio is 31.82, soluble pectin content is 23.72 mg/g, hardness is 1826.55 N, Chewiness is 789.77 mJ and the maximum shear force is 3491.37 N. Correlation analysis showed that there are six key indexes affecting the overall satisfaction of sensory significantly include hardness, chewiness, maximum shear force and physicochemical index of soluble solids, sugar-acid ratio and soluble pectin. The determination coefficient R² of the sensory quality prediction model based on the principal component regression analysis was 0.916, and the standard deviation was 0.11.【Conclusion】The texture and taste quality of pineapple vary greatly among different varieties, and a single evaluation method cannot accurately evaluate its comprehensive quality. , By incorporating sensory evaluation, texture testing and physicochemical analysis, this study identified six key indicators that affecting the overall satisfaction of pineapple, and a prediction model was established for pineapple sensory evaluation  based on the key texture properties and physicochemical indexes, which can accurately predict the texture and taste quality of pineapple, and make up for the lack of objectivity in artificial sensory analysis.

Key words: pineapple, sensory evaluation, texture properties, physiochemical composition, correlation analysis