Automatic ontology evolution in open and dynamic computing environments

Abstract

Automated computing in open and dynamic computing environments requires automatic update and revision of the Knowledge Bases (KBs) to keep the KBs up to date with the dynamics in the environment and correct incorrect knowledge held in the KBs respectively. Furthermore, the truthfulness, applicability and validity of this knowledge depend on the context under which the knowledge is to be used. This then calls for the development of solutions to enable KBs to (i) be evolved over time enabling them to keep up to date with the evolving world or changes in the world's conceptualisation, (ii) allow situational reasoning, and (iii) reasoning under uncertain, incomplete and inconsistent knowledge. The emerging fielding of probabilistic ontologies is impregnated with promises to resolve such issues. However, an investigation on how such knowledge representations can be objectively and rationally evolved is needed. This paper presents issues, methods and ideas towards rational probabilistic ontology evolution in open and dynamic computing environments. © 2010 Springer-Verlag Berlin Heidelberg.