Pattern based mapping and extraction via CIDOC CRM

The current situation within archaeology is one of fragmented datasets and applications, with different terminology systems. The interpretation of a find may not employ the same terms as the underlying dataset. Searchers from different perspectives may not use the same terminology. Separate datasets employ distinct schema for semantically equivalent information. Entities and relationships may have different names but be semantically equivalent. Even when datasets are made available on the Web, effective cross search is hampered by semantic interoperability issues [1]. It is becoming increasingly understood that the use of an integrating conceptual framework, such as the CIDOC Conceptual Reference Model (CRM) (ISO 21127:2006) [2, 21], can help address these issues. We take this as our agreed point of departure. This paper discusses various implementation issues to facilitate use of the CRM. Employing the CRM has tended to require an understanding of the source dataset schema and also specialist knowledge of the CRM and techniques for mappings. This paper argues for the use of mapping patterns to guide deployment, to improve homogeneity, to increase data interchange and to encourage greater uptake.