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

Q&A System Enhancement for Translated Text in Multilingual and Multicultural Context

IP.com Disclosure Number: IPCOM000243608D
Publication Date: 2015-Oct-05
Document File: 2 page(s) / 141K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a Question and Answer (Q&A) system enhancement for translated text in multilingual and multicultural context. This enhancement tool performs a secondary search to provide the most accurate translation from the original text for the user to best understand the results from the primary search.

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

Page 01 of 2

Q&A System Enhancement for Translated Text in Multilingual and Multicultural Context

Today's machine translation technology is mature, but it is never perfect. For example, text translated verbatim from one language to another might not make sense; the terms are correct, but the idea is not correctly expressed. Direct translations do not always make sense.

When the user needs to search data in a Question and Answer (Q&A) system, the search engine searches the text in the same language that matches the search string's language. However, the presented search results may contain text translated from some other language. The translated text might not be completely accurate, compared to the original document. In addition, the translated text might be confusing to the reader.

The current Q&A system has not solved this problem.

The novel contribution is a Q&A system enhancement for translated text in multilingual

and multicultural context. This enhancement is used in a Q&A system. Because the Q&A system can store information in all different languages and the user of the system

can have a different language background, this enhancement provides the most accurate translation from the original text for the user to best understand the search results.

TThe implementation method follows:

1. The system scans the text included in the search results 2. The system analyzes whether the text makes sense in the context when the text is not from the original version, in terms of language 3. If the text is not the original text, then the...