An information flow based approach to semantic integration of distributed digital museums

The distribution of data and knowledge of a huge amount of cultural heritage and the desire to make them available readily to people geographically scattered gives rise to the challenge of semantic integration of distributed digital museums. In order to make this difficult task tractable, how the participant museums are linked conceptually must be looked at first before any implementation issues for links of more a physical nature can be considered. One promising avenue to this end seems to characterize what we call 'informational relationship' between participant museums, which are seen within a (notional) distributed system. And this characterization including reasoning about this relationship should be carried out in both a meaningful and a principled way. In this paper, we describe how to derive semantics in legacy relational database to form local ontology, and present an information flow based approach to mathematically model the relationship among multiple distributed local ontologies in terms of an information channel construction, which involves the construction of classifications and local logics and infomorphisms between them, thus global ontology construction and token connection discovery, To help the establishment of semantic correspondences between various components of the distributed system, which is necessary for the task, the cultural heritage domain ontology CIDOC Conceptual Reference Model (CRM) is deployed as a bridge ontology. The soundness and completeness of the local ontologies is preserved in the global ontology developed in this way.