Implementation of a Referent Tracking System

Traditional database resources and semantic Web technology face problems when there is a need to keep track of individuals in reality as these individuals undergo changes of various sorts. We describe an application which implements the referent tracking paradigm in which each real world entity has its own unique ID. The application is designed to be able to store relationships between tracked instances and also to be extendable to very high orders of magnitude (in principle to accept numbers of entries in the billions). Our approach is based on ontologies grounded in realism, but it can be extended also to information that is captured using the terminologies or concept-based ontologies used in traditional knowledge representation systems. The repository uses resource description framework (RDF) as representation format, and it can thus be queried with query languages such as SPARQL, SeRQL, and RQL, thereby providing support for reasoning over multiple ontologies.