[Corpora-List] English tagger trained on spoken English data

Muhammad Shakir Aziz true.friend2004 at gmail.com
Thu Feb 15 15:02:06 CET 2018


Dear Members Every now and then I need to get some English texts tagged for parts of speech. I am aware that there are very accurate taggers available, for example CLAWS tagger, Stanford tagger, Tree tagger etc. I think that Stanford tagger was trained on Wallstreet Journal texts, i.e. written texts. I am not sure if the other two include models generated on spoken English data. The purpose of this email was to request for your insights in this regard. Spoken language is supposedly different in some ways from written language. Is it ok to use a tagger like Stanford tagger on, for example, conversational texts? Are there any freely available taggers specifically designed for spoken or spoken-like texts (= social media texts, but not limited to tweets)? Thanks for your helpful comments in this regard. Muhammad Shakir PhD Candidate Westfälische Wilhelms-Universität, Münster -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 1022 bytes Desc: not available URL: <https://www.uib.no/mailman/public/corpora/attachments/20180215/5f3e1e10/attachment.txt>


More information about the Corpora mailing list