COMPARISON OF EFFECTIVENESS OF ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM AND ARTIFICIAL NEURAL NETWORKS FOR ESTIMATION OF LINEAR CREEP AND RECOVERY PROPERTIES OF MODEL MEAT EMULSIONS

Abstract

In meat emulsion systems, it is impossible to interpret the results from the molecular perspective due to their complexity. Therefore, it is very difficult to perform mathematical modeling of structure of the emulsion systems due to their ill-defined viscoelastic (creep and recovery) nature. Therefore, an adaptive neuro-fuzzy inference system (ANFIS) was used to accurately model the effect of creep test time, temperature and oil levels on the compliance (J), creep and recovery phase parameters. In this respect, ANFIS with different types of input membership functions (MFs) was developed. MF trimf performed better than others. The ANFIS model was compared with artificial neural network (ANN) and multiple linear regression (MLR) models. The estimation by ANFIS was superior to those obtained by ANN and MLR models. The ANFIS model resulted in a good fit with the observed data, especially for the creep phase data in the checking period. PRACTICAL APPLICATIONS In literature, no study has appeared to model the effect of creep test time, temperature and oil levels on the compliance (J), creep and recovery phase parameters of O/W model system meat emulsions using adaptive neuro-fuzzy inference system (ANFIS), artificial neural network and multiple linear regression techniques. In this study, the ANFIS model resulted in a good fit with the observed data, indicating that the creep and recovery properties of the O/W model system meat emulsions can be estimated using the ANFIS model. As a result, among the models used, ANFIS was found to be the best model that can be efficiently used to estimate unmeasured or untested interval values of creep and recovery properties. This might be quite significant for meat industry that will benefit from estimating texture of such products previously before producing them at a large scale, thus enabling them to save time.