Fuzzy Identification and Modeling of a Gum-Protein Emulsifier in a Model Mayonnaise Color Development System

The aim of this study was to employ iota-carrageenan (IC) and wheat protein (WP) as an emulsifier alternative to egg yolk in a model mayonnaise system. A solution of 0.1% IC and 4% WP was prepared and used as an emulsifier in five different mayonnaise formulas. All mayonnaise treatments were evaluated and compared based on lightness and yellowness (i.e., L and b values respectively) at 4, 23, and 40 degrees C. In addition, an adaptive neuro-fuzzy inference system (ANFIS) was used to model and identify the properties of the resulted mayonnaise, with the temperature and ratios. Experimental validation runs were conducted to compare the measured values and the predicted ones. The L value of the mayonnaise produced from different emulsifiers decreased at the lower storage temperature. The b-value was significantly the highest for mayonnaise formulated from 100% egg yolk. The comparison showed that the adoption of this neuro-fuzzy modeling technique (i.e., ANFIS) achieved a very satisfactory prediction accuracy of about 98%.