Fragflow

FragFlow

Project designed for finding internal macro and composite workflow motifs in scientific workflow, defined according to

http://purl.org/net/wf-motifs#InternalMacro and http://purl.org/net/wf-motifs#CompositeWorkflow.

The project finds a set of workflow fragments from workflow specifications and/or workflow executions and links the results to the corpus. The results are linked according to the Workflow Fragment Description Ontology: http://purl.org/net/wf-fd

In order to achieve the results, this project defines diverse operations for graph manipulation and formatting. In particular:

  • Generic readers and writers that can read and write different workflow specifications and traces (currently supported: OPMW, OPM)
  • Inference and abstraction of a workfow collection or individual workflows.
  • Remote querying and adaptation to process RDF workflows exposed as Linked Data.
  • Formatting output to be read by the SUBDUE and PAFI tools.
  • Capability of saving the results as RDF.
  • Computation of statistics on the results obtained, and binding the fragments proposed by the tools to the results.

The project is configured as a Netbeans project right now. All the libraries and dependencies are jar files contained in the /lib folder.

Current ongoing work:

  • Adapt the framework to different types of graph mining algorithms. Currently supported: SUBDUE, PAFI, Parsemis (gSpan, Gaston(ongoing))
  • Adapt the framework to read from different types of workflows. Currently supported: OPMW, LONI Pipeline
Yolanda Gil
Yolanda Gil
Senior Director for Major Strategic AI and Data Science Initiatives

I have broad research interests in AI and data science. As Principal Scientist I lead the Interactive Knowledge Capture research group, which is part of AI@ISI.