Project Background
With support from two National Science Foundation grants, a team of ASU archaeologists and computer scientists is working to design and build a digital information infrastructure (cyberinfrastructure) for systematically collected archaeological data. Once archaeologists have registered datasets, researchers across scientific disciplines can access these datasets online and issue queries to extract sensibly integrated and appropriately scaled databases of analytically comparable observations from numerous archaeological datasets gathered using incommensurate recording protocols. For archaeology, this project has the goals of (1) advancing archaeologists' ability to engage in synthetic and comparative research and (2) of providing a means to maintain the long-term utility and accessibility of irreplaceable primary data in the face of inadequate metadata and rapidly changing technology. In computer science, the challenges are to develop methods (1) for ad hoc data integration where the semantic demands of the query are reconciled with the semantic content of the available datasets and (2) to resolve conflicts in concept-oriented query processing that arise from inconsistent recording strategies (represented as ontologies).*
The Digital Archaeological Record
The proposed digital information infrastructure is called tDAR for "the Digital Archaeological Record." tDAR is an open-source, Internet accessible foundation for a global archaeological information infrastructure. It is publicly accessible over the Internet, although users must register. tDAR Beta provides basic archival capabilities for databases, spreadsheets, and documents. With the components now under development tDAR will offer concept-oriented queries that do not require knowledge of the schemas of the available datasets and will provide integrated, cross-project, sustainable access to dynamic archives of archaeological, physical anthropological, and environmental data that will:
- advance our ability to conduct synthetic and comparative research;
- and maintain the long-term utility and accessibility of archaeological data.
While we are focusing on systematically collected datasets (for now, databases and spreadsheets), tDAR also archives text and images.
Testbed Project
In order to inform our development and demonstrate the value of the system, knowledge-based data integration tools focusing on archaeological faunal data are being developed under the current grant with the support of a working group of archaeologists with expertise in faunal analysis and the cooperation of the International Council for ArchaeoZoology. (These tools will be readily extensible to other material classes.) The initial testbed project will investigate the socio-environmental conditions that lead to depressed abundance of preferred game over two millennia in the Southwest and lower Illinois River Valleys in the US. So far we have compiled data on 176,000 faunal elements from 81 Southwestern sites.
Computer Science Challenges
An ultimate integrated view of multiple data sets is usually impossible and unnecessary. Thus, we reconcile the semantic demands of a query with the semantic content of the available datasets (rather than attempting global reconciliation of data sources). tDAR uses a novel strategy of query-driven, ad-hoc data integration in which, given a query, the cybertools will identify relevant data sources and perform interactive, on-the-fly metadata matching to align key portions of the data while reasoning with potentially incomplete and inconsistent information.
Grant Support
Archaeological Data Integration for the Study of Long-term Human and Social Dynamics 2006 National Science Foundation, Human Social Dynamics Grant 0624341 (3 years)
Principal Investigators: Keith W. Kintigh, K. Selçuk Candan, Hasan Davulcu, Subbarao Kambhampati, Margaret C. Nelson, Katherine A. Spielmann
Key ASU Collaborators: Huiping Cao, John Howard, Allen Lee, Ben Nelson, and Yan Qi.
Enabling the Study of Long-Term Human and Social Dynamics: A Cyberinfrastructure for Archaeology 2004 National Science Foundation, Human Social Dynamics Grant 0433959 (1 year)
Principal Investigators: Keith W. Kintigh, John M. Anderies, Chitta R. Baral, K. Selçuk Candan, Hasan Davulcu, Michelle Hegmon, Subbarao Kambhampati, Ann Kinzig, Huan Liu, Peter H. McCartney, Ben Nelson, Margaret C. Nelson, Charles L. Redman, Arleyn W. Simon, Katherine A. Spielmann, and Sander van der Leeuw
This initial NSF grant funded a December 2004 workshop that provided a vision for the infrastructure and suggested a number of design requirements. The workshop's full report was published in American Antiquity(July 2006) and has been endorsed by the Society for American Archaeology, the Society for Historical Archaeology, and the American Association of Physical Anthropologists. The workshop included 31 individuals including archaeologists with diverse interests, computer scientists, and domain scientists associated with other science informatics projects.
Digital Antiquity: Planning a Digital Information Infrastructure for Archaeology 2007 Andrew W. Mellon Foundation, Scholarly Communications Program (1 Year).
Principal Investigators: Keith Kintigh (Arizona State University), Jeff Altschul(SRI Foundation), Tim Kohler (Washington State U.), Fred Limp (U. of Arkansas), Julian Richards (U. of York) and Dean Snow (Penn State U.)
This grant from the Andrew W. Mellon Foundation led to the creation of archaeoinformatics.org, a partner organization to tDAR. This organization is following up the NSF workshop recommendations to develop a sociologically attractive, technologically feasible, and financially sustainable plan for the organization and operation of a cyberinfrastructure for archaeology.
