Hello,
I have recently developed a module that performs SQL alike queries, and RegExp alike pattern matching on natural language data with syntactic annotations.
It is intended to query data with syntactic questions such as
find all the common nouns that are attached to a verb as subjects and that have at least one child that is a preposition.
or find all the occurrences of pairs (Verb + Oblique case with the preposition 'as') on a set of sentences.
The code is not yet open and I'm still writing some unit tests, but I have written some basic documentation that can be found here.
I would like to know if you believe these kind of functionalities could be integrated into NLTK?
I see there are some modules for information extraction, and I believe this module could be a good add
Thank you very much!
Gabriel