A community approach to data integration: Authorship and building meaningful links across diverse archaeological data sets

The ability to link and compare diverse archaeological data sets will catalyze innovative research of great scope and analytic rigor. However, information heterogeneity and limited budgets and information technology skills challenge data dissemination initiatives. This paper argues for new methods of community-based data integration pioneered by the University of Chicago’s Extensible Markup Language (XML) System for Textual and Archaeological Research project (XSTAR). With XSTAR, data integration takes place in two steps: (1) syntactic-schematic integration: Legacy data sets are migrated for representation in the data structures described by the Archaeological Markup Language (ArchaeoML), and (2) Semantic integration: Mappings must be established between related terms and classes in each source database. Because the nuances of meaning are often very subtle, human experts must classify related items in each data set. Initial syntacticschematic mapping of data into XSTAR is simple and fast but occurs at a relatively