Found this, and I don’t know what I’d prefer for my use case.
Hi Daniel,
Enabling question semantics causes the system to assume all text in the corpus consists of questions (queries), and to encode them in a more specialized way.
A query against such a corpus will return the most similar queries to the one being asked. Two common use cases are:
- Searching the titles of FAQs for the question most similar to the one being asked.
- Searching a preexisting query set for similar queries with known good answers.
However, if your content source contains a mix of question and non-question material, the default setting will work better.
I hope this helps.
1 Like