It seems that such corpora are best discussed at Linguistics.SE and questions about specific NLP frameworks such as NLTK are best reserved for SO. And yet these questions have been asked:

I can see how finding existing corpora could be on topic here, but I'm not sure questions about how to construct a sentiment dictionary should be on topic. What do you think?


If questions about how to generate new open data are not a good fit for Open Data, then I'm honestly not sure what the point of this site is other than to ask for existing data sets, which doesn't seem like a very constructive use for a Stack Exchange site.

Such questions are not welcome on Stack Overflow.

I don't really see why such questions would be a better fit for Linguistics, though there is a "corpora" tag there. It seems to me that corpora-generation has more to do with open data than with the study of linguistics.


I think you have two different kinds of examples here. I agree that the first is off topic, and I see it's been closed already. I definitely think that questions about the existence of and where to find linguistic data are at least as on topic as questions about hardware GPU data. (I flagged that GPU one as off-topic, but then read the comments and I'm willing to revoke that flagging, at least in spirit. I don't see technically a way to retract it.)

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .