Ontology & Triples

IMO, you'd better use the dependency parsing technique instead of the one you're doing now which is called the constituency parsing. This way you'd find it quite straightforward to get the triples from a sentence. Say, I have the following sentence:

Bell, based in Los Angeles, makes and distributes electronic, computer and building products.

The dependency graph for the sentence is the following:

nsubj(makes-8, Bell-1)
nsubj(distributes-10, Bell-1)
partmod(Bell-1, based-3)
nn(Angeles-6, Los-5)
prep_in(based-3, Angeles-6)
conj_and(makes-8, distributes-10)
amod(products-16, electronic-11)
conj_and(electronic-11, computer-13)
amod(products-16, computer-13)
conj_and(electronic-11, building-15)
amod(products-16, building-15)
dobj(makes-8, products-16)
dobj(distributes-10, products-16)

And guess what? Yes, the results are already in triples. You just need to annotate them with terms from some NLP (Natural Language Processing) ontology.

References:

1. Constituency vs Dependency Parsing
2. Stanford typed dependencies manual
3. Stanford Parser