COMPARISON OF AN ADAPTIVE RESONANCE THEORY-BASED NEURAL-NETWORK (ART-2A) AGAINST OTHER CLASSIFIERS FOR RAPID SORTING OF POST CONSUMER PLASTICS BY REMOTE NEAR-INFRARED SPECTROSCOPIC SENSING USING AN INGAAS DIODE-ARRAY

An Adaptive Resonance Theory Based Artificial Neural Network (ART-2a) has been compared with Multilayer Feedforward Backpropagation of Error Neural Networks (MLF-BP) and with the SIMCA classifier. Al three clas sifiers were applied to achieve rapid sorting of post-consumer plastic s by remote near-infrared (NIR) spectroscopy. A new semiconductor diod e array detector based on InGaAs technology has been experimentally te sted for measuring the NIR spectra. It has been found by a cross valid ation scheme that MLF-BP networks show a slightly better discriminatio n power than ART-2a networks. Both types of artificial neural networks perform significantly better than the SIMCA method. A median sorting purity of better than 98% can be guaranteed for non-black plastics. Mo re than 75 samples per second can be identified by the combination InG aAs diode array/neural network. However, MLF-BP neural networks can de finitely not extrapolate. Uninterpretable predictions were observed in case of test samples that truly belong to a particular class but that are located outside the subspace defined by training set.