Data processing by neural networks in quantitative chemical analysis

Abstract

An overview is given of the current usage of artificial neural networks as mathematical models for non-linear calibration procedures. The emphasis is on practical aspects: the choice of the calibration samples, the required network characteristics for a given problem, various training methods and their efficiency and the validation of the network models. Some problems with the application of neural networks in multivariate calibration are considered, together with recent research aimed at solving these problems