Formal concept analysis with background knowledge: a case study in paleobiological taxonomy of belemnites
Abstrakt
We present a case study in identification of taxa in paleobiological data. Our approach utilizes formal concept analysis and is based on conceiving a taxon as a group of individuals sharing a collection of attributes.
In addition to the incidence relation between individuals and their attributes, the method uses expert background knowledge regarding importance of attributes which helps to filter out correctly formed but paleobiologically irrelevant taxa. We present results of experiments carried out with belemnitesa group of extinct cephalopods which seems particularly suitable for such a purpose.
We demonstrate that the methods are capable of revealing taxa and relationships among them that are relevant from a paleobiological point of view.