Structural Theory of Science as a Systematic Framework for the Design of DL's and CD's for E-science.
Hansje Braam · Description Logics · 2007
The construction of ontologies (DL-instances, Concrete Domains) for E-Science (in my individual case E-Humanities) is quite hard: i) in many cases we have to cope with ontological divergencies, up to completely disjunct ontologies, as many competing theories (and scientific practices) rely on non-common features, entity-classes and relations; ii) many scientifically relevant predicates have arity > 2, a potential problem for DL(CD)’s; iii) in science (natural science, but humanities as well) formalisms are used with high complexity (mathematics, formal grammars, predicate logics, restricted natural language etc.); iv) higher order predicates occur -especially in the humanities(f. e. it is possible to reason about certains modes of interpretation). It is a.o. this semantical complexity which distinguishes science from folk cognition. I propose to use Structural Theory of Science (the classical reference [1], see also [2]) as a common framework for the analysis of scientific theories (and practices) in which the specifications of adequate ontologies can be related and subsequently used in the systematic construction of the DL’s and CD’s.