Incorporating semantics in scientific workflow authoring

Abstract

Abstract. The tools used to analyze scientific data are often distinct from those used to archive, retrieve, and query data. A scientific workflow environment, however, allows one to seamlessly combine these functions within the same application. This increase in capability is accompanied by an increase in complexity, especially in workflow tools like Kepler, which target multiple science domains including ecology, geology, oceanography, physics, and biology. To overcome this complexity, we have developed semantically-driven user-interface components that are customized at run-time using domain-specific ontologies. One such subsystem in Kepler uses domain-specific ontologies to customize the presentation of analytical components and data for use by scientists building workflows. Kepler also provides for semantically-enabled queries for components, which can significantly increase efficiency in workflow authoring tasks. In this demonstration, we show how ontologies can be used for user-interface customization and more. In particular, we show our recent ontology-driven extensions for workflow authoring in Kepler. These extensions include our advances in: (1) automating dataintegration and service-composition tasks, (2) the use of semantic annotations to verify that workflows are semantically meaningful, and (3) the ability to search for contextually relevant components and data sets in situ, i.e., as a user is designing a scientific workflow. 1.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 67 citations exist upstream.