Sztuczne sieci neuronowe Kohonena jako narzędzie w taksonomii paleontologicznej - metodyka oraz zastosowanie na przykładzie późnokredowych belemnitów

Artificial neural networks (ANNs), the computer software or systems that are able to "learn" on the basis of previously collected input data sets are proposed here as a new useful tool in paleontological modeling. Initially ANNs were designed to imitate the structure and function of natural neural systems such as the human brain. They are commonly used in many natural researches such as physics, geophysics, chemistry, biology, applied ecology etc. Special emphasis is put on the Kohonen self-organizing mapping algorithm, used in unsupervised networks for ordination purposes. The application of ANNs for paleontology is exemplified by study of Late Cretaceous belemnites. The Kohonen networks objectively subdivided the belemnite material] ~ 750 specimens) into consistent groups that could be treated as monospecific. The possibility of transferring these results to the language of classical statistics is also presented. Further development and possibility of use of ANNs in various areas of paleontology, paleobiology and paleoecology is briefly discussed.