Semantic Workflows
I have become very active in a new area: the use of
AI techniques
to support scientific analysis.
Our work has focused on semantic workflows that describe
the input data, computations in the workflow steps, and all results of the workflow execution
using semantic web languages (OWL, RDF, SPARQL) which are W3C standards. We have developed
expressive representations of workflows, as well as a variety of reasoning algorithms for
workflow composition through interactive assistance, workflow validation,
automated workflow completion, metadata propagation, and
workflow retrieval.
A major result from our work is the Wings workflow system.
On the Wings site
there are publications,
a web-accessible installation of our workflow system with a tutorial,
examples from different science domains,
and
open source software if you are interested in downloading it.
See also the Pegasus project site.
Semantic Workflows in Wings
These are some suggested readings about Wings, more articles are available in the
Wings publications site.
- A good introductory overview of how users can interact with WINGS:
- A brief overview of the workflow reasoning algorithms in WINGS:
- A formal description of the WINGS workflow representation language and detailed
workflow reasoning algorithms:
Other relevant pointers are:
- Workflow provenance: Recording
and sharing both the executed workflow and its reusable abstract description.
- Workflow Design: Principles
for the design of complex software applications as workflows.
- Workflow Sharing: Reuse,
provenance, and usability of workflows across user groups and communities.
- CAT: an intelligent assistant
for workflow composition.
General References on AI and Scientific Workflows
A good introduction and overview of this general area is:.
Another good introduction to scientific workflows is in the
Final Report of the
NSF Workshop on Challenges of Scientific Workflows
that I recently co-chaired, summarized in this article:
-
Examining the Challenges of Scientific Workflows,
Yolanda Gil, Ewa Deelman, Mark Ellisman, Thomas Fahringer, Geoffrey Fox, Dennis Gannon,
Carole Goble, Miron Livny, Luc Moreau, and Jim Myers.
IEEE Computer, vol. 40, no. 12, pp. 24-32, December, 2007.
-
Wings for Pegasus: Creating Large-Scale Scientific Applications
Using Semantic Representations of Computational Workflows,
Yolanda Gil, Varun Ratnakar, Ewa Deelman, Gaurang Mehta, and Jihie Kim.
Proceedings of the 19th Annual Conference
on Innovative Applications of Artificial Intelligence (IAAI),
Vancouver, British Columbia, Canada, July 22-26, 2007.
Slides from the presentation.
I gave a tutorial on
"Computational Workflows for Large-Scale AI Research",
at AAAI-08, ask me for the slides if you are interested.
Check out also a talk that I gave at the US National Science Foundation
titled
Artificial Intelligence and Cyberinfrastructure:
Workflow Planning and Beyond.
There is also material and slides from a panel I organized
at
IAAI-04
on
The Broader Role of Artificial Intelligence in Large-Scale Science Research.
I co-edited a special issue of
IEEE Intelligent Systems on e-science, together with
David DeRoure and Jim Hendler, in January 2004.
The Guest Editor's Introduction and Table of Contents are available on-line.
Also take a look at my presentations and other materials available
from a
Dagstuhl seminar on Semantic Grid held on July 2005.