Artificial neural network approach to the evaluation of the coordination geometry in organotin(IV) compounds

Abstract

Artificial neural networks (ANNs) are a simple and rapid system for pattern recognition. In this study they were used to classify Mössbauer spectra of penta-coordinated and octahedral Sn(iv) complexes. Mössbauer spectra recognition is a lengthy procedure requiring a great deal of experience. The application of a system such as artificial neural networks provides a rapid and accurate method for the correct classification of Mössbauer spectra. As the two categories of spectra are not linearly separable, conventional techniques like principal component analysis (PCA) or perceptron can not be used. A more complex ANN was therefore used to solve this problem. The network was built as a standard three-layer back-propagation network with 256 input neurons, 2 hidden neurons and 1 output neuron and a sigmoidal activation function. The network's performance was tested with test sets of 10, 20 and 50% of the total number of spectra. The mean square error (MSE) of the different test sets show significant differences. This type of network was able to classify correctly the spectra with an MSE of less than 0.030. Moreover, the network was even able to classify in the appropriate class a spectrum that had been deliberately inverted, demonstrating the ability of ANN to recognize objects affected by noise or distortion.

Quantitative determination of Freon gas using an electronic nose

“…The ANN used in this work consists of three layer feed-forward ANN, 11,13 as shown in Fig. 1, consisting of input layer, hidden layer and output layer.…”