Ontology Classifications

1. Improving Simulation Management Systems through Ontology Generation and Utilization Jonathan P. Leidig, Edward A. Fox, Kevin Hall, Madhav Marathe, Henning Mortveit Contact: leidig@vt.edu Abstract Ontology Classifications Content from simulation systems is useful in defining domain ontologies. We describe a digital library process to generate and leverage domain ontologies to support simulation systems tasks. Workflow ontologies may be used to define compositions of simulation-related services. Simulation model ontologies may be used in customizing collection management systems for tasks such as organization, interface construction, and metadata record generation. Simulation model ontology Domain-specific meta-ontology Context-specific ontology Language support Compatible analyses Result schema Model ontology relationships (e.g., malaria,influenza) • Context ontology • relationships • (e.g., epidemiology, network science) Validation Inputschema • Context ontology Model ontology • Context ontology Model ontology • Context ontology Model ontology Targeted Simulation Systems • Epidemiology Applications • Malaria models • Influenza models • ODE and agent-based models • Models from NIH MIDAS community • Models from Gates Foundation community • Analysis applications • Network analysis • Model-specific analysis • Digital Library Integration • Institutional infrastructure • Network science cyberinfrastructure Ontology Generation and Technologies • Ontology Formats • XML schema • RDF • Ontology Generation • Human-intensive model ontology generation • Metadata description set generation software • Harmonization yields context-specific ontologies • Harmonization • RDF descriptions • Software guided humanmapping • Ontology Terms • Dublin Core terms • Infrastructure and collection-level terms • 5S framework terms • Model and context-specific terms Recommending and Selecting Model Ontology Harmonization Context Ontology Harmonization Simulation Workflows Simulation Process Domain Meta Ontologies Analysis Process Model-Specific Ontologies Context-Specific Ontologies Sample Content Input Files Result Files Input Configuration Output Result Dataset Documentation Analysis Annotation Schema Result Summaries Analyses Experiment Products Example Records (XML, RDF) DB Metadata Schemas (DDL) Model-Specific Description Sets Prototype Implementation & Applications Supported Harmonized Description Sets • Swiss Tropical Institute • Malaria models • Dataset analysis • Cyberinfrastructure Network Science • Network simulations • Network analysis • Virginia Bioinformatics Institute • Biological domains • Infectious diseases (e.g., H1N1, H5N1) • Biological organs • Infrastructure domains • Transportation systems • Computer and wireless networks Model Ontology-Utilizing Digital Library Services • Content staging • Interface presentation of model parameters • Input parameter gathering • Input configuration generation • Input configuration validation • Input, result, and analysis storing and retrieving • Gathering provenance from workflow stages • Model-specific indexing • Faceted browsing • Ranked searching Acknowledgement This work has been partially supported by NSF SDCI Grant OCI-1032677, NSF Nets Grant CNS-062694, CNS-0831633, HSD Grant SES-0729441, CDC Center of Excellence in Public Health Informatics Grant 2506055-01, NIH-NIGMS MIDAS GM070694-05/06, and DTRA CNIMS Grant HDTRA1-07-C-0113. Related Article: Jonathan Leidig, Edward Fox, Kevin Hall, Madhav Marathe, Henning Mortveit. SimDL: A Model Ontology Driven Digital Library for Simulation Systems. ACM/IEEE Joint Conference on Digital Libraries, Ottawa, Canada, June 13-17, 2011.