The problem of Question Answering (QA) as used in TREC [
13] can be formulated as follows:
Given a collection of natural-language (NL) documents find an answer to given NL query that is a short substring of one of
the documents, and it is found in a relevant context.
We present a novel approach to the problem based on unification-based grammars for NL.
Keywords question answering - stochastic unification-based grammars
This work is supported by NSERC and OGS.