Data Model

Each statement in CS-KG refers to a specific claim extracted from one or more research articles in the form <subject, predicate, object>.

This schema exemplifies the structure of a typical statement in CS-KG.
The examplary statement refers to the triple <cskg:human_face_detection, skos:broader, cskg:computer_vision> extracted from 2 papers. The triple is reified using the properties rdf:subject, rdf:predicate, and rdf:object. The statement has type cskg-ont:Statement and provo:Entity (the PROV Ontology is used to track the provenance of a claim).
The number of articles from where the triple was generated is expressed by the datatype property cskg-ont:hasSupport. The provenance is provided by linking the statement to the modules of the pipeline that generated the triple (object property provo:wasGeneratedBy) and the papers where it was extracted from (object property provo:wasDerivedFrom).

This simple structure can be used the query the CS-KG via the SPARQL endpoint. For instance, the following query retrieves all the statement about the research entity "sentiment analysis" (ordered by number of associated papers).

                    PREFIX rdf:  <http://www.w3.org/1999/02/22-rdf-syntax-ns#> 
                    PREFIX cskg: <https://w3id.org/cskg/resource/> # CS-KG resources 
                    PREFIX cskg-ont: <https://w3id.org/cskg/ontology#> # CS-KG ontology 

                    SELECT (cskg:sentiment_analysis as ?sub) ?prop ?obj ?sup FROM <https://w3id.org/cskg>
                    WHERE {
                        ?t rdf:subject cskg:sentiment_analysis ;
                        rdf:predicate ?prop ;
                        rdf:object ?obj ;
                        cskg-ont:hasSupport ?sup .
                    }
                    ORDER BY desc (?sup)
                    LIMIT 100
                    	

The "help" section in the SPARQL endpoint offers a set of examplary queries as a starting point.

Ontology

The CS-KG ontology is available here as OWL file.
The full documentation of the ontology is available here.

For information and questions please contact:
Danilo Dessì – danilo [dot] dessi [at] unica [dot] it
Francesco Osborne – francesco [dot] osborne [at] open [dot] ac [dot] uk