A tool to compute and visualize the Graph of Rules Dependencies of an ontology composed of existential rules with default negation. It can compute a stratification if the rule base is stratifiable and apply forward chaining on a given fact base.
Classical usage :
java -jar graal-stratified-negation.jar [options]
List of supported options :
-f,--input-fileRule set input file. Default: --r,--rule-setPrint the rule set. Default: false-g,--grdPrint the Graph of Rule Dependencies. Default: false-s,--print-sccPrint the Strongly Connected Components. Default: false-G,--print-gsccPrint the graph of the GRD Strongly Connected Components. Default: false-c,--forward-chainingApply forward chaining on the specified Fact Base. Default: --h,--helpPrint this message. Default: false-v,--versionPrint version information Default: false-w,--windowLaunch the GUI. Default: false
To launch the GUI use :
java -jar graal-stratified-negation.jar -w
A user manual is available in doc/manual.pdf. The Graphical User Interface provides the same functionalities as the command line usage, even more.
- Some colors :
- Green : the green color (on a rule or a strongly connected component) means that everything is OK
- Red : the red color means that there is a problem and the Ontology can not be stratified
- On a rule : the rule belongs to a circuit wich contains a negative reliance
- On a strongly connected component : the strongly connected component is a circuit wich contains a negative reliance -While displaying the Graph of Strongly Connected Components, clicking on a strongly connected component will open a new window with the Graph Of Rule Dependencies associated to it.
- You can export an ontology in DLGP format
- You can export a saturated fact base in DLGP format
- You can export both the Graph of Rule Dependencies and the Graph of the Strongly Connected Components
In order to work with bigger ontologies this tool needs some improvments :
- Import Rules from a database
- Import Facts from a database
- Export saturated set of facts to a database