13 Ontologies and Knowledge-Based Systems

The most serious problems standing in the way of developing an adequate theory of computation are as much ontological as they are semantical. It is not that the semantic problems go away; they remain as challenging as ever. It is just that they are joined - on center stage, as it were - by even more demanding problems of ontology.

- Smith (1996) p. 14

How do you go about representing knowledge about a world so it is easy to acquire, debug, maintain, communicate, share, and reason with? This chapter explores how to specify the meaning of symbols in intelligent agents, how to use the meaning for knowledge-based debugging and explanation, and, finally, how an agent can represent its own reasoning and how this may be used to build knowledge-based systems. As Smith points out in the quote above, the problems of ontology are central for building intelligent computational agents.