Browse Prior Art Database

Machine translation system that uses Japanese conversion rules in Japanese generation processing

IP.com Disclosure Number: IPCOM000012255D
Original Publication Date: 2003-Apr-23
Included in the Prior Art Database: 2003-Apr-23
Document File: 2 page(s) / 19K

Publishing Venue

IBM

Abstract

At the last stage of Japanese generation, the machine translation system uses Japanese to Japanese conversion rule to make natural Japanese translation result.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 57% of the total text.

Page 1 of 2

THIS COPY WAS MADE FROM AN INTERNAL IBM DOCUMENT AND NOT FROM THE PUBLISHED BOOK

JP820021098 Koichiro Kato/Japan/IBM Tomohiro Miyahira, Harumi Itoh

  Machine translation system that uses Japanese conversion rules in Japanese generation processing

Main point of this invention Disclosed is an English to Japanese machine translation system which can generate natural Japanese translation sentences by applying conversion rules of Japanese -> Japanese, in the last stage of the machine translation system.

Background of the invention In the pattern-base machine translation system, syntactic analysis of an input text and syntactic generation of a result translation are simultaneously performed using synchronous tree adjoining grammar. The phrase structures of the result translation are determined when phrase structures of corresponding input text are determined. Therefore, it was difficult to change a part of translation result according to the case whether the sentence is negative or affirmation. For example, when the part of "either a pen or a pencil" in the following text is analyzed as a noun phrase, the structure of translation result is determined as a noun phrase "

".

Input text: He has either a pen or a pencil.

Translation result:

Therefore, even if the input sentence is changed to a negative sentence, it is translated as the following incorrect result.

Input text: He doesn't have either a pen or a pencil.

Translation result:

Detailed description of the invention
1....