Pattern recognition methods for classification of soils based on their radionuclide content

Abstract

Multivariate data analysis methods were used to recognize and classify soils of unknown geographic origin. A total of 103 soil samples were differentiated into classes, according to regions in Serbia and Montenegro from which they were collected. Their radionuclide (226Ra, 238U, 235U, 40K, 134Cs, 137Cs, 232Th and 7 Be) activities detected by gamma-ray spectrometry were then used as the inputs in different pattern recognition methods. The prediction ability of linear discriminant analysis (LDA), k-nearest neighbours (KNN), soft independent modelling of class analogy (SIMCA) and artificial neural network (ANN) were 82.8%, 88.6%, 60.0% and 92.1%, respectively.