36E-57


Modeling of fruit leather development by cluster analysis

A. Jangchud1, S. VATTANAKUL1, R. Attabhanyo2, and S. Sampanvejsobha3. (1) Dept. of Product Development, Faculty of Agro-Industry, Kasetsart Univ., 50 Phaholyothin Rd, Chatuchak, Bangkok, 10900, Thailand, (2) Dept. of Food Science and Technology, Chiang Mai Univ., Chiang Mai, Thailand, (3) Dept. of Food Science and Technology, Mae Fa Luang Univ., Chiang Rai, Thailand

Fruit leather could be made from different varieties of fruit. Several ingredients have been attempted and experimented for the best formula optimization for each fruit. The process of formulation in fruit leather product development could be shorten by using modeling technique. The objective of this study was to develop model for fruit leather formulation development by cluster analysis technique. Seven tropical fruits were selected for this study: lynchee, rambutan, longan, mango, guava, pineapple and orange. The chemical properties such as total soluble solid, pectin, pH, total acidity, total sugar, reducing sugar and crude fiber were analyzed. Ingredients named TextureAid-S, TextureAid-A, TextureAid-B, TextureAid-C, TextureAid-D, TextureAid-E, TextureAid-F, TextureAid-G and TextureAid-H were used to develop in each variety of fruit at different concentrations and different combinations. Sensory evaluation has been conducted by an untrained panel using hedonic scale. Finished products were also analyzed on physical, chemical, and sensorial properties. The data on chemical composition and optimum ingredients were analyzed and modeled using cluster analysis technique. Results from cluster analysis showed that five clusters of fruit leather formulation were constructed according to their chemical composition and ingredients used. Pectin content, total soluble solid and crude fiber were the most suitable properties on cluster analysis. Example showed that mango had high pectin and no TextureAid-F added in the formulation while guava had low pectin that TextureAid-F needed in the formulation. The constructed cluster of fruit could be applied on unknown fruit leather development by identify its cluster by using chemical properties and select ingredient according their groups.

Session 36E, Fruit & Vegetable Products: General
8:30 AM - 12:00 PM, Monday AM Room Hall I-2

2005 IFT Annual Meeting, July 15-20 - New Orleans, Louisiana