Neural Fuzzy Network Model with Evolutionary Learning Algorithm for Mycological Study of Foodborne Fungi

This study developed a neural fuzzy network (NFN) model with evolutionary learning algorithm for use in the field of food mycology for predicting growth in foodborne fungi. The evolutionary learning algorithm in the proposed model is a hybrid Taguchi-genetic algorithm (HTGA) that simultaneously finds the optimal antecedent and consequent parameters by directly minimizing root-mean-squared error (RMSE), which is a key performance criterion. The minimum RMSE is then used to optimize the number of fuzzy rules for the NFN. Experimental results show that the proposed HTGA-based NFN model with eight fuzzy rules outperforms recently reported neural networks in terms of accuracy in predicting the maximum specific growth rate of foodborne Monascus ruber.