PortageLive: delivering machine translation technology via virtualization

18288

Authors

nan

Abstract

This paper is a case study of the deployment of new MT technology aiming at improving the overall Post-Editing (PE) experience. One of the main challenges in having MT output post-edited within a localization workflow is being able to meet ever increasing quality ex- pectations, especially from a terminology perspective. In Symantec‟s localization process we constantly seek to overcome this obstacle by refining our MT output in a num- ber of ways. Several layers of customization have therefore been added over the years to our commercial MT system (such as User Dictionaries, Translation Stylesheets and Au- tomated Post-Editing). Despite obtaining sub- stantial quality gains with these techniques, improvements are still sought to minimize PE effort. The deployment of a novel technology, based on SYSTRAN's new hybrid approach, is presented in this paper. This paper focuses on the technical and linguistic challenges as- sociated with the integration of this new tech- nology into an existing MT workflow. 1

Keywords

nan

Full Content

Deploying novel MT technology to raise the bar for quality: A review of key advantages and challenges Johann Roturier Symantec Ballycoolin Business Park Blanchardstown, Dublin 15, Ireland Johann_roturier@symantec.com Abstract This paper is a case study of the deployment of new MT technology aiming at improving the overall Post-Editing (PE) experience. One of the main challenges in having MT output post-edited within a localization workflow is being able to meet ever increasing quality ex- pectations, especially from a terminology perspective. In Symantec‟s localization process we constantly seek to overcome this obstacle by refining our MT output in a num- ber of ways. Several layers of customization have therefore been added over the years to our commercial MT system (such as User Dictionaries, Translation Stylesheets and Au- tomated Post-Editing). Despite obtaining sub- stantial quality gains with these techniques, improvements are still sought to minimize PE effort. The deployment of a novel technology, based on SYSTRAN's new hybrid approach, is presented in this paper. This paper focuses on the technical and linguistic challenges as- sociated with the integration of this new tech- nology into an existing MT workflow. 1 Introduction In Symantec's Localization department we have been using MT in conjunction with TM technology to translate product documentation in some of our localization workflows for several years. This ap- proach has been mostly based on an in-house op- timization of a customized SYSTRAN MT system and on the post-editing of MT output by external vendors. Using MT has allowed us to increase the speed and consistency of translation, which in turn also enabled us to process larger document sets within shorter turnaround times. This paper is di- vided into three main sections. Section 2 focuses on the criteria we have been using for evaluating and deploying MT technology. Section 3 focuses on the customization techniques that we have been using to improve the quality of our MT output. Finally, section 4 presents the findings of a recent pilot project that was conducted to assess the dep- loyment of a novel MT component, based on SY- STRAN's new hybrid approach. The advantages of this component will be demonstrated and the chal- lenges encountered during the deployment of this new technology will also be discussed. 2 Defining criteria for evaluating and dep- loying MT technology During our initial review of MT systems, we iden- tified a number of criteria to evaluate and deploy MT technology. When new MT components be- come available, such as the one based on SYSTRAN‟s hybrid approach, we re-use these cri- teria. 2.1 Identifying the right team While it is often suggested that (S)MT systems do not require linguists to be built, the same statement does not apply when it comes to deploying and maintaining a production MT system, especially when this system needs to be updated over-time. Evaluating the improvements achieved by a given MT system can be done semi-automatically but human checks are necessary and even essential to make sure that MT resources do not become obso- lete. Deploying …

Content truncated to 500 words

Labeled Text

This paper is a case study of the deployment of new MT technology aiming at improving the overall Post-Editing (PE) experience. One of the main challenges in having MT output post-edited within a localization workflow is being able to meet ever increasing quality expectations, especially from a terminology perspective. In Symantec‟s localization process we constantly seek to overcome this obstacle by refining our MT output in a number of ways. Several layers of customization have therefore been added over the years to our commercial MT system (such as User Dictionaries, Translation Stylesheets and Automated Post-Editing). Despite obtaining substantial quality gains with these techniques, improvements are still sought to minimize PE effort. The deployment of a novel technology, based on SYSTRAN's new hybrid approach, is presented in this paper. This paper focuses on the technical and linguistic challenges associated with the integration of this new technology into an existing MT workflow. 1

Publication Details

Paper ID:
18288
Year:
2009
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan