Question Validation

• I have searched both the documentation and discord for an answer.

Question

To facilitate the creation of a knowledge graph validation module, I'd like to obtain the original text chunk for a given triplet.

Currently, it appears that index_struct only associates the chunks at keyword granularity, which isn't enough to identify the originating chunk for a given triplet.

Let's say, for example, I have a triplet (wolves, are, carnivorous). KGIndex currently will only allow me to find every textnode chunk related to the subject "wolves" - which is not useful in a corpus that probably contains many references to this subject across chunks.

How can I identify the originating textnode for a given full triplet (subject, predicate, object)? Is this possible?