The potential for computer-aided identification in biodiversity research

Abstract

Species identification lies at the heart of most ecological studies, but it is recognized as a difficult and often frustrating task. Taxonomists have sought to overcome the difficulties of species identification by developing a range of tools and techniques that have increasingly involved the use of computers. We describe recent developments in computer-aided species identification, which have been in four main areas; multi-access keys; hypertext keys; expert systems; and neural networks. We also suggest which technique might be appropriate for a particular taxonomic group.