https://github.com/dgarijo/fragflow

Project desinged for detecting common fragments in scientific workflows by reusing several existing graph mining techniques
https://github.com/dgarijo/fragflow

Last synced: 28 days ago
JSON representation

Project desinged for detecting common fragments in scientific workflows by reusing several existing graph mining techniques

• Host: GitHub
• URL: https://github.com/dgarijo/fragflow
• Owner: dgarijo
• Created: 2013-10-30T17:11:07.000Z (about 11 years ago)
• Default Branch: master
• Last Pushed: 2015-07-28T09:40:20.000Z (over 9 years ago)
• Last Synced: 2024-04-16T19:32:56.976Z (9 months ago)
• Language: GLSL
• Homepage:
• Size: 44.2 MB
• Stars: 2
• Watchers: 3
• Forks: 1
• Open Issues: 47
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
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