Genetic implementation of a classifier based on data separation by means of hyperspheres

This paper discusses a genetic implementation of the growing hyperspheres classifier (GHS) for high-dimensional data classification. The main idea of the GHS classifier consists in data separation by n-dimensional hyperspheres properly spread over the training data. First, the idea of training data representation is described. Then a brief description of a previous first representation by neural networks is reminded. The main part of this paper is focused on a precise description of the classifier implementation by genetic algorithms. Features of the new approach are discussed and compared. Finally, a task classifying data from a gamma telescope is presented to show the capabilities of the classifier.