eyeprolog

RDF 1.2 triple terms in a named graph

Does a claim keep its context when it is unpacked?

Alice’s claim that Bob knows Carol is stored in a named graph of claims. EyeProlog lifts the quoted relationship out of the triple term and keeps the graph it came from, so the result never loses its context.

Run it in the playground · Source

---

The question

Which relationship does the claim contain, and in which graph does the result belong?

This is one of a family of RDF 1.2 examples. RDF data arrives as rdf(Subject, Predicate, Object, Graph) facts, as produced by rdf-prolog-interchange. The fourth argument is the graph: default_graph, or the name of a graph in a TriG dataset.

---

The data

rdf(iri('https://example.org/alice'), iri('https://example.org/claims'), triple(iri('https://example.org/bob'), iri('https://example.org/knows'), iri('https://example.org/carol')), iri('https://example.org/claimsGraph')).

TriG is the RDF syntax for datasets with several named graphs. A named graph groups triples under a name, which is often used to record where they came from or how much they are trusted.

---

The rule

result_rdf(S, iri('https://example.org/knows'), O, G) :-
  rdf(
    _,
    iri('https://example.org/claims'),
    triple(S, iri('https://example.org/knows'), O),
    G
  ).

The same variable G appears in the source pattern and in the result, so the derived relationship lands in the graph that supplied the claim.

---

The answer

eyeprolog examples/rdf12-trig-triple-term.pl

result_rdf(iri('https://example.org/bob'), iri('https://example.org/knows'), iri('https://example.org/carol'), iri('https://example.org/claimsGraph')).

The result is in claimsGraph, not in the default graph.

---

The reason

step(result_rdf(iri('https://example.org/bob'), iri('https://example.org/knows'), iri('https://example.org/carol'), iri('https://example.org/claimsGraph')),
     rule(2),
     ['S' = iri('https://example.org/bob'),
      'O' = iri('https://example.org/carol'),
      'G' = iri('https://example.org/claimsGraph')],
     [rdf(iri('https://example.org/alice'), iri('https://example.org/claims'), triple(iri('https://example.org/bob'), iri('https://example.org/knows'), iri('https://example.org/carol')), iri('https://example.org/claimsGraph'))]).

The binding 'G' = iri('https://example.org/claimsGraph') shows where the graph name was carried through.

---

The check

condition('C1', resolution, ok, 2).
condition('C4', coverage, ok, 2).
condition('C7', relevance, ok, 3).
trusted(0).
verdict(checked).

---

Try it

• Replace G in the rule head with default_graph and decide whether that loses information a consumer might need.
• Add the same claim in a second graph and predict the number of results.
• Compare with the default-graph version and with named-graph reasoning.

Learn more in The Art of EyeProlog, Chapter 15.

This site is open source. Improve this page.