Provenance Data Models and Assertions: A Demonstrative Approach
113
5.6 ROV-DC
PROV-DC (Dublin Core architecture) enables the conversion of provenance from the
DC vocabulary [31] to the PROV-DM vocabulary. This signifies that an application
which supports PROV can easily consume provenance already exposed as Dublin
Core. DC is used for multiple reasons ranging from a simple description of a resource
to combining metadata vocabularies besides providing interoperability for metadata
in the Semantic Web as many of its terms are used for expressing provenance.
5.7 ProvONE
Provenance models proposed so far have been the ones that provide for database
provenance. These models aim to support the authenticity and attribution of a data
item. Thus, such kind of models is suitable for tracking the lineage of material objects
like artworks, manuscripts, database flows, etc. In scientific experiments, however,
we require shreds of evidence in the support of experimental results. The data in these
scientific workflows is obtained by the means of computational results. That is in the
form of structured graphs specific to each computational step. The provenance information thus provided is encoded by the means of machine processes and is used for
query and analysis. Workflow tracking systems like Taverna [32], Kepler [33], and
VisTrails [34] collect this information using the proprietary models. These systems
also adopt varied models to specify workflows. This creates a system of heterogeneity which makes it difficult for scientists to analyze and compare provenance
information generated from different workflows that are generated or enacted using
different systems. This in turn results in missed opportunities for stitching the traces
provided by different workflows [35]. ProvONE data model was thus proposed to
fulfill the above-mentioned requirements and provide a standard format for modeling
provenance with regards to scientific workflow systems. ProvONE is a model that
will work on the top of the existing PROV-DM model and provides for procedures
that easily amalgamate with the existing PROV-DM data model [36].
6 Tools for Capturing Provenance
6.1 Karma
Karma [37] is a tool that is widely used in cyberinfrastructure for collecting and maintaining provenance records. Karma’s modular architecture allows for the multiple
instrumentation plug-ins that allow for its use in different architectural settings.
Karma provenance management tool allows users to easily capture provenance
information inside scientific workflows.
113
5.6 ROV-DC
PROV-DC (Dublin Core architecture) enables the conversion of provenance from the
DC vocabulary [31] to the PROV-DM vocabulary. This signifies that an application
which supports PROV can easily consume provenance already exposed as Dublin
Core. DC is used for multiple reasons ranging from a simple description of a resource
to combining metadata vocabularies besides providing interoperability for metadata
in the Semantic Web as many of its terms are used for expressing provenance.
5.7 ProvONE
Provenance models proposed so far have been the ones that provide for database
provenance. These models aim to support the authenticity and attribution of a data
item. Thus, such kind of models is suitable for tracking the lineage of material objects
like artworks, manuscripts, database flows, etc. In scientific experiments, however,
we require shreds of evidence in the support of experimental results. The data in these
scientific workflows is obtained by the means of computational results. That is in the
form of structured graphs specific to each computational step. The provenance information thus provided is encoded by the means of machine processes and is used for
query and analysis. Workflow tracking systems like Taverna [32], Kepler [33], and
VisTrails [34] collect this information using the proprietary models. These systems
also adopt varied models to specify workflows. This creates a system of heterogeneity which makes it difficult for scientists to analyze and compare provenance
information generated from different workflows that are generated or enacted using
different systems. This in turn results in missed opportunities for stitching the traces
provided by different workflows [35]. ProvONE data model was thus proposed to
fulfill the above-mentioned requirements and provide a standard format for modeling
provenance with regards to scientific workflow systems. ProvONE is a model that
will work on the top of the existing PROV-DM model and provides for procedures
that easily amalgamate with the existing PROV-DM data model [36].
6 Tools for Capturing Provenance
6.1 Karma
Karma [37] is a tool that is widely used in cyberinfrastructure for collecting and maintaining provenance records. Karma’s modular architecture allows for the multiple
instrumentation plug-ins that allow for its use in different architectural settings.
Karma provenance management tool allows users to easily capture provenance
information inside scientific workflows.