Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Part-Of-Speech Assumption of Japanese Lemma-Form Word

IP.com Disclosure Number: IPCOM000116285D
Original Publication Date: 1995-Aug-01
Included in the Prior Art Database: 2005-Mar-30
Document File: 4 page(s) / 112K

Publishing Venue

IBM

Related People

Ogino, S: AUTHOR

Abstract

Disclosed is a system which estimates Part-Of-Speech (POS) of the stem of an input Japanese word. This system consists of one more content word dictionaries, a POS estimation module, and an optional-feature setting module. The set of POS which is used for the estimation is: {Wa-gyou-5dan (Wa-5) verb (V), Ka-gyou-5dan-1 (Ka-5-1) V, Ka-gyou-5dan-2 (Ka-5-2) V, Ga-gyou-5dan (Ga-5) V,Sa-gyou-5dan (Sa-5) V, Ta-gyou-5dan (Ta-5) V, Na-gyou-5dan (Na-5) V, Ba-gyou-5dan (Ba-5) V, Ma-gyou-5dan (Ma-5) V, Ra-gyou-5dan (Ra-5) V, SURU-type (SuruN) verbal-noun (V,N), Suru-special-conjugation (Suru) V, Zuru-special-conjugation (Zuru) V, Ka-gyou-special-conjugation (Ka-S) V, Ra-gyou-special-conjugation (Ra-S) V, Adjective (Adj), Adjective noun (AdjNoun), Noun, Suffix, Adverb (Adv), Prefix.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 65% of the total text.

Part-Of-Speech Assumption of Japanese Lemma-Form Word

      Disclosed is a system which estimates Part-Of-Speech (POS) of
the stem of an input Japanese word.  This system consists of one more
content word dictionaries, a POS estimation module, and an
optional-feature setting module.  The set of POS which is used for
the estimation is:
  {Wa-gyou-5dan (Wa-5) verb (V), Ka-gyou-5dan-1 (Ka-5-1) V,
   Ka-gyou-5dan-2 (Ka-5-2) V,
  Ga-gyou-5dan (Ga-5) V,Sa-gyou-5dan (Sa-5) V, Ta-gyou-5dan (Ta-5) V,
  Na-gyou-5dan (Na-5) V, Ba-gyou-5dan (Ba-5) V, Ma-gyou-5dan (Ma-5)
V,
  Ra-gyou-5dan (Ra-5) V, SURU-type (SuruN) verbal-noun (V,N),
  Suru-special-conjugation (Suru) V, Zuru-special-conjugation (Zuru)
V,
  Ka-gyou-special-conjugation (Ka-S) V, Ra-gyou-special-conjugation
   (Ra-S) V,
  Adjective (Adj), Adjective noun (AdjNoun), Noun, Suffix, Adverb
   (Adv), Prefix.

      Ka-5-1 and Ka-5-2 are different in renyo-form connected to
particle "te".  The inflection of Ka-5-1 renyo-form is "i", and the
one of Ka-5-2 is small "tsu".

      When the input word is in the system dictionary, POS of the
word in the dictionary is considererd the most presumable.  The
result of the estimation is consequently placed after POS set found
in the dictionary.

      The different two levels for POS estimation accuracy can be
specified with the parameter for accuracy-level setting.  The default
value of the parameter is set to be the lower level.  In the lower
level esti...