Quality prediction of mangosteen during storage using artificial neural network (IAEC Ref. 163)

Abstract

In this research, multilayer perceptron neural network with back propagation algorithm was applied to predict mangosteen quality during storage at the most appropriate pre-storage conditions which performed the longest storage period. Each condition was combination of three pre-storage treatments involved pre-cooling at 20oC, bee waxing concentration of 6% and strecth film single wrapping. To decide which combination gave best fruit quality on their prolonged storage, it was investigated by panelist preference used hedonic scale. Based on the experiment, pre-cooling at 20oC followed by bee waxing concentration of 6% and strecth film single wrapping provided minimum quality changes over 40d. Those optimum data combination was used to predict quality changes during storage using neural network. The model was successfully trained used network architecture at 30,000 iterations; 14 nodes in hidden layer; learning rate constant 0.8 and momentum 0.7. It was represented by coefficient correlation (R2) closed to 1 (more than 0.99) for each parameter, indicated that model was good to memorize data yet R2 validation was poor. It defined that network architecture developed quality prediction model had not appropriated yet to predict mangosteen quality during storage using new data variants.