Multilayer Perception Artificial Neural Network and Radial Basis Networks as Tools to Predict Shelf Life of Cheese Tikka
Abstract: Soft cheese is well known acid coagulated dairy product. “Tikka” is a general South Asian term meaning marinated barbequed food. Cheese Tikka, which is a very popular vegetarian delightful, exotic dish, is traditionally made by inserting pieces of marinated soft cheese pieces in between grills and barbequed for several minutes over hot tandoor (oven). It is served with tomato sauce and spices, and known as a delightful cuisine. Cheese tikka´ requires over five hours for its preparation. Feedforward and radial basis models were developed to predict shelf life of cheese tikka, and compared with each other. Input parameters were moisture, titratable acidity, free fatty acids, and tyrosine content. Overall acceptability score was taken as output parameter. The dataset were portioned as 90.47% for training, and 9.53% for testing. Several important aspects, for the first time have been highlighted and compared for predicting shelf life of cheese tikka. Different combinations of several internal parameters, i.e., data pre-processing, data partitioning, number of hidden layers, number of neurons in each hidden layer, transfer function, error goal, spread constant etc., along with backpropagation algorithm based on bayesian regularization mechanism as training function, and sum square error as performance function were used during training of the neural network. The network was trained with 100 epochs. The number of neurons in each hidden layer varied from 1 to 30. Experimentally obtained shelf life of cheese tikka is 40 days. Feedforward backpropagation soft computing model gave better results as evidenced from root mean square error (RMSE 7.39%) than radial basis function soft computing model (8.65% RMSE). The feedforward backpropagation soft computing model detected 38.87 days, which is comparable to experimentally discovered shelf life of 40 days.
Full-Text Available: PDF
Authors: Submit Goyal, Gyanendra Kumar Goyal