Browse Prior Art Database

Proofreading assistance for translation

IP.com Disclosure Number: IPCOM000022457D
Original Publication Date: 2004-Mar-16
Included in the Prior Art Database: 2004-Mar-16
Document File: 3 page(s) / 66K

Publishing Venue

IBM

Abstract

A program is disclosed that presents reviewers with locations in the translated document of possible mistranslation, omission or wrong terms. The mechanism is based on identifying words that are missing in translation, using the English-Japanese dictionary prepared for translators.

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 99% of the total text.

Page 1 of 3

Proofreading assistance for translation

[Framework] The figure shows the mechanism employed for the tool. For each set of the source and translated sentences (or segments), the program performs the following verifications:
- Breaks down the source sentence into words and determines if the each word is found in the dictionary (A).

- Normalizes the word form (converting words from uppercase to lowercase, from plural to singular form, and from past to present tense) for comparison with the dictionary
(B). - Obtains a list of translated words for the source word, if it is found in the dictionary (C).

- Searches the translated segment for the string that matches the word found on the list (D).

- Concludes that there is a high probability of translation error, if the translation does not include a string that matches any of the words listed from the dictionary (E). - Presents the source and translated segments with the translated words provided from the list (F).

[Examples]
(1) Example of translation without mistranslation, omission or wrong words

1

[This page contains 1 picture or other non-text object]

Page 2 of 3

(2) Example of mistranslation

2

[This page contains 1 picture or other non-text object]

Page 3 of 3

[Advantages] - The program does not require its own data, since it uses the existing dictionary made available to translators.

- The analysis is easy, reliable and fast, only requiring word division and string matching.

3

[This page contains 1 picture or other...