A Multi-Lingual Web-Based Survey Form Machine Translation Mechanism

Document Type

Conference Proceeding

Publisher

Australian Undergraduate Students' Computing Conference

Faculty

Faculty of Computing, Health and Science

School

School of Computer and Security Science

RAS ID

2636

Comments

Boddington, R. , Clayden, J. M., Collins, M. , & Pride, S. (2003). A Multi-Lingual Web-Based Survey Form Machine Translation Mechanism. Proceedings of First Australian Undergraduate Students' Computing Conference. (pp. 13-17). Melbourne, Australia. Australian Undergraduate Students' Computing Conference. Conference proceedings available here.

Abstract

Translation costs restrict the preparation of medical survey questionnaires in migrant and Aboriginal communities in Western Australia. This is further compounded by a lack of affordable and accurate machine translation (MT) mechanisms. This research investigates combined strategies intended to provide an efficacious and affordable machine translator to meet these needs by: (a) using an interlingua or hub-language which requires less resources for its construction than contemporary systems and has the additional benefit of significant error reduction; (b) creating word group structures to reduce the complexity of translation rules and enhance correct transfer of meaning between natural languages; and (c) defining smaller operating environments to restrict data input and further enhance translation by reducing error occurrence significantly. This research aims to produce a prototype MT mechanism that will accept questionnaire texts as discrete questions and suggested answers from which a respondent may select. The prototype will be des igned to accept as input, non-ambiguous English as the source language (SL), translate it to a 'hub -language' or interlingua based on Esperanto and thence to a selected target language (TL). Additionally, an inverse path of translation from French back to English will enable validation of minimal or zero change in both syntax and semantics of the original input.

Access Rights

Free_to_read

Share

 
COinS