There are free or very affordable corpora: Ontonotes and GUM, for example. Cheers,
R On Sat, Dec 16, 2017 at 11:46 PM, Benedict Holland <[email protected]> wrote: > No. It isn't free. This is how linguists make money. That said, the data > isn't expensive. I think the name training data is less than 1,000 dollars. > It might be less for academic use. > > Thanks, > ~Ben > > On Dec 16, 2017 2:37 PM, "Jeff Zemerick" <[email protected]> wrote: > >> Unfortunately, I don't think that data is available. >> >> Jeff >> >> >> On Fri, Dec 15, 2017 at 1:33 AM, SAGAR MEHAR <[email protected]> >> wrote: >> >> > Dear All, >> > Thanks for the Information, Jeff. I tried training my own model >> but I >> > have very less training data available. Is it possible to get the >> training >> > data which was used to make en-ner-person model. Is this data available >> for >> > free without any restrictions? >> > >>
