Title:
ORTHOGRAPHICAL VARIANT DETECTION DEVICE AND ORTHOGRAPHICAL VARIANT DETECTION PROGRAM
Document Type and Number:
WIPO Patent Application WO/2012/169128
Kind Code:
A1
Abstract:
Provided is an orthographical variant detection device with which an orthographical variant candidate is detected with good precision. An orthographical variant detection device of an embodiment comprises: a vocabulary extraction unit which extracts vocabulary from document data; a similarity computation unit which computes the similarity of an arbitrary pair of terms of the extracted vocabulary; an orthographical variant candidate assessment unit which assesses whether the pair of vocabulary terms is an orthographical variant candidate on the basis of the similarity; and a group classification unit which groups the orthographical variant candidate on the basis of text character strings which are included in both terms in the orthographical variant vocabulary term pair which is the orthographical variant candidate.
Inventors:
KOBAYASHI MIKITO
SHINOZAKI MASAHISA
KANO TOSHIYUKI
SAITO YOSHIMI
SHINOZAKI MASAHISA
KANO TOSHIYUKI
SAITO YOSHIMI
Application Number:
PCT/JP2012/003357
Publication Date:
December 13, 2012
Filing Date:
May 23, 2012
Export Citation:
Assignee:
TOSHIBA KK (JP)
TOSHIBA SOLUTIONS CORP (JP)
KOBAYASHI MIKITO
SHINOZAKI MASAHISA
KANO TOSHIYUKI
SAITO YOSHIMI
TOSHIBA SOLUTIONS CORP (JP)
KOBAYASHI MIKITO
SHINOZAKI MASAHISA
KANO TOSHIYUKI
SAITO YOSHIMI
International Classes:
G06F17/21; G06F17/30; G06F40/00
Foreign References:
| JPH07225763A | 1995-08-22 | |||
| JP2007179505A | 2007-07-12 |
Attorney, Agent or Firm:
SAGOI, MASAYUKI (JP)
Masayuki Sunai (JP)
Masayuki Sunai (JP)
Download PDF:
Claims: