Browse Prior Art Database

System and method for semi-real time evaluation of the applicability of machine translation based on context-sensitive social polling

IP.com Disclosure Number: IPCOM000241555D
Publication Date: 2015-May-11
Document File: 4 page(s) / 98K

Publishing Venue

The IP.com Prior Art Database

Abstract

The proposed invention provides the confidence level of machine translation to users by combining machine translation and social polling. The native speakers can give instant feedback such as "Yes" or "No" for evaluating the contents from machine translation based on the context sensitive input data like audios, pictures or videos in addition to texts.

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

Page 01 of 4

System and method for semi

System and method for semi-

--real time evaluation of the applicability of machine translation based on

real time evaluation of the applicability of machine translation based on

real time evaluation of the applicability of machine translation based on

context

context-

--sensitive social polling

sensitive social polling

It's a globalization world. In order to speed up communication, translation demand has been increasing at a rapid speed, Aside from the two known approaches: Human Translation and Machine Translation, nowadays, social network translation is also very popular - users post translation questions in the community and wait native speakers to answer and feedback. With assistance from native speakers, users can can get high quality results, however, the response time is still a question mark, since it depends on a variety of factors.

The present invention proposed a method to shorten the waiting time of social translation. It utilized MT and broadcast MT output to native speakers. The evaluation mechanism is easy and simple so that more native speakers are willing to feedback as it only takes them a few seconds. Based on the feedback, the system then generates a confidence level of the MT output for reference.

The proposed invention provides the confidence level of machine translation to users by combining machine translation and social polling. The native speakers can give instant feedback such as "Yes" or "No" for evaluating the contents from machine translation based on the context sensitive input data like audios, pictures or videos in addition to texts.

The primary advantage of the proposed invention is to utilize social network power and machine translation to get semi-real time trustworthy translation service.

The proposed invention can be implemented as a mobile application that incorporates a back-end machine translation engine, a database of social communi...