rdf-cmap

RDFCMap converts between CMap CXL and RDF TTL.

Usage to convert from CXL to TTL

c:\path\to\jdk8\bin\java -Dlog4j.configurationFile=resources/log4j2.xml -jar rdf-cmap.jar -i your-input-cmap-model.cxl

Note: Conversion from CXL to TTL supports vocabulary of AFO2.0 as well as version CR/2018/07

Usage to convert from TTL to CXL:

c:\path\to\jdk8\bin\java -Dlog4j.configurationFile=resources/log4j2.xml -jar rdf-cmap.jar -i your-input-rdf-model.ttl

Usage to visualize data description of ADF as CXL:

c:\path\to\jdk8\bin\java -Dlog4j.configurationFile=resources/log4j2.xml -jar rdf-cmap.jar -i your-input-adf-file.adf

Note: Creation of a useful visualizations of large graphs requires automatic layouting of nodes. Rdfcmap supports layouting based on graphviz as well as different layout algorithms provided by gephi. Please contact OSTHUS for further information (office(at)osthus(dot)com).

Use --help to see command line options c:\path\to\jdk8\bin\java -Dlog4j.configurationFile=resources/log4j2.xml -jar rdf-cmap.jar --help

How to create a cmap to be used with rdfcmap?

1. Get CMAP Tools from https://cmap.ihmc.us/
2. Create a new cmap following guidelines below on formatting, see documentation of ADM for details about formatting of Allotrope data models
3. All nodes of a cmap model are transformed to instances in rdf model
4. Labels of cmap nodes are transformed to titles of instances using dct:title
5. Links between nodes in cmap are transformed to object properties in rdf model.
6. Labels of links must agree to existing labels (skos:prefLabel) of properties in RDF vocabulary.
7. Cardinalities can be added to object properties in cmap: default without specification of cardinality is "min 0". Exact cardinality is specified by adding the number, e.g. "has part =1" means there is exactly 1 link of "has part". Minimum cardinality is specified by ">1" (one or more)
8. Export cmap as CXL file

How to create a SPARQL query?

1. Prepare a CMAP
2. Change style of your root node to oval with dashed border.
3. Change style of your target node to oval with solid border.
   • Each SPARQL query is created as a graph starting from a specified root node to a specified target node.
   • By default all values of properties of the target node will be returned as query result
4. Delete all nodes from your cmap that do not need to be included in your query graph
   • Keep all nodes on the path from root node to target node
   • Keep further nodes if required for disambiguation
5. Save CMAP and export as CXL
6. Execute rdfcmap with following command line parameters

• c:\path\to\jdk8\bin\java -Dlog4j.configurationFile=resources/log4j2.xml -jar rdf-cmap.jar -i your-cropped-query-graph-from-cmap.cxl -r vocabulary.ttl --sparql --notitles --droplongcomments
• vocabulary.ttl is an optionally specified file containing vocabulary that helps to create better understandable queries

3. Resulting SPARQL query is created next to your input file.

Changelog:

V2.4.2

• updated creation of SPARQL to support LC-UV model CR/2018/07

V2.4.1

• fixed IRIs of CHEBI instances

V2.4.0

• improved processing of labels
• support LC-UV model CR/2018/07

V2.3.4

• improved sparql export
• improved handling of user-specified prefix mapping

V2.3.2

• Create visualization directly from ADF
• Note: Reading and writing ADF requires presence of ADF-API in build path. Please contact OSTHUS for further support and information (office(at)osthus(dot)com).

V2.3.1

• Improved conversion from ttl to cmap in order to support blank nodes and literal values
• Improved import of triples
• Note: Conversion from TTL to CXL requires layouting of concepts in order to get a useful visualization. Enabling automatic layouting with rdfcmap needs additional third-party dependencies. Currently, rdfcmap supports layouting based on graphviz as well as different layout algorithms provided by gephi. Please contact OSTHUS for further information (office(at)osthus(dot)com).

V2.2.0

• Updated conversion from ttl to cmap in order to support BFO aligned models
• Note: Conversion from TTL to CXL requires layouting of concepts in order to get a useful visualization. Enabling automatic layouting with rdfcmap needs additional third-party dependencies. Currently, rdfcmap supports layouting based on graphviz as well as different layout algorithms provided by gephi. Please contact OSTHUS for further information (office(at)osthus(dot)com).

V2.1.4

• Updated conversion from cmap to ttl in order to support BFO aligned models
• Added SPARQL export
• Added Shape export
• Added ADF export

©OSTHUS 2016-2018