Prediction and sensitivity analysis by TS fuzzy neural network for fungal growth in food products

Abstract

A TS fuzzy neural network (TSFNN) was compared with an artificial neural network (ANN) in terms of accuracy in predicting the combined effects of temperature, pH level, sodium chloride level and sodium nitrite level on the growth rate of Leuconostoc mesenteroides. The TSFNN and ANN models were compared in terms of six statistical indices calculated by comparing their prediction results with actual data. The learning-based systems obtained encouraging prediction results. Sensitivity analyses of the four environmental factors showed that temperature and, to a lesser extent, NaCl had the most influence on accuracy in predicting the growth rate of Leuconostoc mesenteroides. The observed effectiveness of TSFNN for modeling microbial kinetic parameters confirms its potential use as a supplemental tool in predictive mycology. Comparisons of the six statistical indices showed that the TSFNN model was better than ANN model in predicting the four kinetic parameters.