Hubble classification of galaxies using neural networks

It is shown that artificial neural networks have the ability to classify galaxies into the (Hubble) classes elliptical, spiral, irregular and to some degree lenticular. Most misclassification occurs between neighbouring classes on the Hubble sequence. The results show that correlations exist between the morphological type of a galaxy and measurable galaxy features, indicating that it is possible to define the Hubble sequence in terms of specific galaxy parameters.