Application of FA-ANN to Discriminate Tea Varieties Based on Spectroscopy

Pattern recognition problems specifically for spectral data were developed. As an application, classification of four tea varieties based on near infrared spectra was taken by using the method. Factor analysis (FA) and artificial neural networks (ANN) were used for pattern recognition in this research. FA is a very effective data mining way; it was applied to enhance species features and reduce data dimensionality. ANN with back propagation algorithm was used for the data compression tasks as well as class discrimination tasks. The first 6 principal components computed by FA were applied as inputs to a back propagation neural network with one hidden layer. This model can correctly recognize all the 100 samples in the calibration set. This model was used to predict the variety of 20 unknown samples. The recognition rate of the model for the unknown sample is 100%. So this paper could offer an effective discrimination way