Semantic Integration for Marine Science Interoperability Using Web Technologies

Citation Information

Title: Semantic Integration for Marine Science Interoperability Using Web Technologies
Authors: Rueda, C., Bermudez, L., Graybeal, J., Isenor, A., and Alexander, P
Conference Information: AGU Fall 2008 Meeting
Date: 2008.12.19
Location: San Francisco, CA.  USA.
Paper number: IN53A-1186

Abstract

The Marine Metadata Interoperability Project, MMI (http://marinemetadata.org), promotes the exchange, integration, and use of marine data through enhanced data publishing, discovery, documentation, and accessibility.  A key effort of the project is the definition of an Architectural Framework and Operational Concept for Semantic Interoperability, which is complemented with the development of tools that realize critical uses cases in semantic interoperability.

This poster describes the overall Semantic Framework architecture and its main operational components, exemplified with applications in the context of data management challenges in the marine and environmental sciences. The components are Semantic Web tools that perform important interoperability tasks, ranging from the creation of controlled vocabularies and the mapping of terms across multiple ontologies, to the online registration, storage, and search services needed to work with the ontologies (http://mmisw.org). This set of services uses Web standards and technologies, including Resource Description Framework (RDF), Web Ontology language (OWL),  Web services, and toolkits for Rich Internet Application development.

The MMI Ontology Registry and Repository is the central component of our architecture. It provides registry and storage services for ontologies, which are complemented with various related functionalities, including interfaces to create vocabularies, upload existing ontologies, capture relevant metadata to better serve the submitted documents, and support sophisticated search functions. Users and client applications can submit search requests using the W3C SPARQL Query Language for RDF.  An important feature is the ability to directly resolve the Uniform Resource Identifiers (URI) associated with the registered ontologies as well as their concepts via HTTP requests.  This is provided by a URI Resolver component. 

The Voc2RDF service converts a comma- or tab-delimited set of terms and definitions into an RDF file.  This service greatly facilitates the creation of controlled vocabularies by using a simple and intuitive form-based user interface. Created vocabularies and their descriptive metadata can be seamlessly submitted to the MMI Ontology Registry for versioning and community access.

The Vocabulary Integration Environment tool, VINE, allows the user to map vocabulary terms across multiple ontologies. Various relations can be established between the terms, for example exact, narrow, and related matching relations, as defined by the Simple Knowledge Organization System specifications, (http://www.w3.org/2004/02/skos/), but others can be added. Attributes about each mapping, like comments and a confidence level, can also be included.  VINE is integrated with the MMI Registry and Repository and can also be used as an standalone application.