Browse Prior Art Database

Partial Translation of Capitalized Words in Machine Translation

IP.com Disclosure Number: IPCOM000123980D
Original Publication Date: 1999-Sep-01
Included in the Prior Art Database: 2005-Apr-05
Document File: 2 page(s) / 51K

Publishing Venue

IBM

Related People

Harada, M: AUTHOR

Abstract

This article describes a method to find out a noun phrase referring capitalization of words for better machine translation process.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Partial Translation of Capitalized Words in Machine Translation

This article describes a method to find out a noun phrase
referring capitalization of words for better machine translation
process.

   General writing rule in English is to write words with
lower case characters except the first character of a sentence and I.

   If there is a word which starts with uppercase character
(here, the word is called "capitalized word") or  a sequence of
capitalized word (here, the word sequence is called "capitalized
word sequence"), it may mean:
  - Proper Noun
  - Title
  - Product Name

   These types of words or word sequences are to be handled
as a noun phrase.

   If a preposition is included in a title or a product name,
the preposition can be written either in upper case or lower case.

   In addition, version, edition, or model names are often
found after the word sequence should be included as whole.

   As the result, word sequence defined with following figures
are to be detected and to be handled as a noun phrase in Machine
Translation process.
     +<--------------------------------------------+
   |   | START ->+-> Capitalized word ->+--->+-> Preposition ->+
   |                      |   |
   +<---------------------+   +->+-> Version description ->+
                          |   |                         |
                          |  +-> Edition description ->+
                          |   |                         |
                         ...