
The novel unfolds on a 10x10 grid, where each square (cell) is a place and a chapter. Each chapter tells the story of a main character. The sequence of the cells is predefined and based on the chess knight piece move, with the requirement that it cannot step two times on the same cell. Cell A1 is skipped.
Classes:
Data Properties:
Object Properties:
For this step we identified the ODP “Sequence” from ontologydesignpatterns.org:
http://ontologydesignpatterns.org/wiki/Submissions:Sequence
Even if we have a connection between cells and chapters we can use to determine the sequence of cells (given that the sequence of chapters is incremental), we integrated this ontology for faster data extraction.
We asserted only the property directlyFollows to define the sequence of the cells, leaving the reasoner to infer all the other properties (follows, precedes, directlyPrecedes).
The consistency of the model has been checked with the Pellet reasoner, while SPARQL queries with SPARQL Query tab in Protege.
In order to keep the graph light and Protege working, we put in the ABox only the data needed to answer the Competency Questions for the final tests.
That is, while we entered all the Cells and the Chapters, only a few Places and Characters were put in the graph.
2022, Diego Chillo & Laura Travaglini