USE OF ARTIFICIAL NEURAL NETWORKS FOR MODELING SURVIVAL OF Alicyclobacillus acidoterrestris CRA7152 IN APPLE JUICE
Artificial Neural Networks (ANN) were proposed as an alternative technique in the
field of predictive microbiology. This study was based on the survival determination of
A. acidoterrestris CRA 7152 in apple juice under the effects of pH (3.5 to 5.5),
temperature (25 to 50°C), nisin concentration (0 to 70IU/ml) and soluble solids-Brix
(11 to 19). A quadratic polynomial was developed and used as comparison to the
neural model. An artificial network with 4 nodes in the input layer, 4 nodes in the
hidden layer and 1 node in the output layer gave the best fit to the studied process.
The values of the statistical indices for model validation Mean-Relative-PercentageResidual and Mean-Absolute-Relative-Residual were -17.43 and 17.81 respectively for the polynomial model and, -1.67 and 14.54 for the neural model, showing that
ANN performed slightly better than the polynomial model, allowing thus the use of this technique in microbial modeling.
Este trabalho está licenciado sob uma licença Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.