A FRAMEWORK TO MANAGE INFORMATION MODELS – THE PLANETARY DATA SYSTEM CASE STUDY. J. S. Hughes1, D. J. Crichton1 and C. A. Mattmann1, 1Jet Propulsion Laboratory (4800 Oak Grove Dr. Pasadena, CA 91109, {[email protected], [email protected], [email protected]}). Introduction: The Planetary Data System (PDS) Information Model has been formally captured in a tool framework based on an ontology modeling tool. Originally developed in the late 1980’s, the informa-tion model had become inconsistent in many areas causing significant problems in data collection, prepa-ration and validation. A PDS team determined the most reasonable interpretations of the model from ex-isting documentation and developed an ontology. A specification document, which can be automatically generated from the framework, now provides a basis for streamlining, simplifying and improving the PDS standards for use both within PDS and internationally. This is critical as PDS begins a major upgrade towards the next generation of its standards. Information Model: The information model is the foundation on which an information system is built. It defines the entities to be processed, their attributes, and the relationships that add meaning. It also provides the source for the descriptors, taxonomies, and classi-fication schemes required by science information sys-tems to meet their requirements. The development and subsequent management of the information model is the single most significant factor in the development of a successful information system. Information Model Management: The tool framework supports the management of an information model with the rigor typically afforded to software development. This framework provides for evolution-ary and collaborative development independent of sys-tem implementation choices. Once captured, the mod-eling information can be exported to various languages and notations for documentation and semiautonomous generation of code for traditional and semantic web applications. This framework is being successfully used for several science information systems including the Planetary Data System (PDS) [1,5], the Interna-tional Planetary Data Alliance (IPDA) [2], the Na-tional Cancer Institute’s Early Detection Research Network (EDRN) [3], and several Consultative Com-mittee for Space Data Systems (CCSDS) projects. The PDS Information Model: The PDS archives and distributes scientific data from NASA planetary missions, astronomical observations, and laboratory measurements. The PDS information model - also known as a metadata model – defines the data, mis-sions, instruments, repositories, and people that are part of the planetary science domain. The defined con-cepts and terms allow people and computers to accu-rately communicate and subsequently support data storage, archive, search, locate, retrieval, processing, and reasoning. Ontology Modeling Tool: The core component of the framework is an ontology modeling tool [4,6] into which the information model is captured. The content can be exported to semantic web languages such as the Web Ontology Language (OWL) and Re-source Description Framework (RDF). The content can also be written to XML Metadata Interchange (XMI) files for import into UML tools allowing the generation of code and the expression of the model in a variety of notations including concept maps and class diagrams. Figure 1 illustrates a portion of the PDS conceptual model using a concept map. Figure 1 – Example Concept Map Significance: The tool framework supports a model driven architecture. It formally captures the information model and allows the content to be used directly in the development process. This approach is expected to make the PDS more agile as it supports the continually changing Planetary Science domain. References: [1] J. S. Hughes, et al., "An Ontology-Based Archive Information Model for the Planetary Science Community," In Proc. Spaceops, Heidelberg, Germany, 2008. [2] Hughes, J.S., Crichton, D.J., Beebe, R., Preliminary Definition of the Core Archive Data Standards of the International Planetary Data Alliance (IPDA), PV 2007 International Conference, http://pv2007.dlr .de/Papers/Hughes_CoreDataStd.pdf, 2007. [3] J. S. Hughes, D. Stelling, D.Crichton, C. Mattmann, G. Warnick, S. Reid. An Information Model for Biomarker Research. In Proceedings of the 5th EDRN Scientific Workshop, Bethesda, MD,March 17-19, 2008. [4] H.Wache, T. Vogele, U. Vis-ser, H. Stuckenschmidt,G. Schuster, H. Neumann and S. Hubner , Ontology-Based Integration of Information — A Survey of Existing Approaches, Proceedings of IJCAI-01 Workshop:Ontologies and Information Shar-ing, Seattle, WA, 2001, Vol. pp. 108-117. [5] J. S. Hughes and S. K. McMahon, "The Planetary Data Sys-tem. A Case Study in the Development and Manage-ment of Meta-Data for a Scientific Digital Library," Research and Advanced Technology for Digital Li-braries, Springer Berlin/Heidelberg, January 1998. [6] Protégé Ontology Editor and Knowledge-Base Frame-work, Stanford Center for Biomedical Informatics Re-search, http://protege.stanford.edu/,