1

For example the word to be lemmatized is "managerial" or "financial". But lemmatization does not support grouping of managerial as "manager" or financial as "finance". What alternative can be used to get the root word(noun) from an adjective?

s111
  • 34
  • 3
  • Are you looking for the same behavior of [this question](http://stackoverflow.com/questions/36042494/extract-main-word-from-its-descendants-with-python/36042629#36042629) ? – Till Apr 04 '16 at 08:50
  • yes.I am doing the text categorization.So i need to fine tune the transcript [get the rootwords] and then apply text categorization – s111 Apr 04 '16 at 08:52
  • `managerial -> manager` is not exactly lemmatization. Lemmatizer shouldn't change the POS, you're looking for something loosely term as "linguistic regularlities", e.g. http://msr-waypoint.com/en-us/um/people/gzweig/Pubs/NAACL2013Regularities.pdf and http://www.aclweb.org/anthology/W14-1618 – alvas Apr 04 '16 at 08:56
  • Thank you. Is there a way to get the noun for an adjective? – s111 Apr 04 '16 at 09:04
  • I think the "state-of-art" now is to do some vector models using neural nets. like the 2 papers I've linked in the previous comments. No easy way out =( – alvas Apr 04 '16 at 09:18

0 Answers0