Work Packages


  • WP1 is concerned with planning, organising, directing and controlling the total effort to efficiently achieve FAUST's objectives. General management will be the responsibility of UCAM.


  • The development of an architecture that will permit the evaluation and exploitation of novel web-oriented, feedback collection mechanisms that reduce noise in feedback provided by users and increase the utility of the web contributions.
  • The creation of mechanisms for instantaneous incorporation of user feedback into 24x7 operational translation engines.
  • The creation of seamless flows and bridges between the production, research, and evaluation environments.


  • Creation of and with interfaces and support for MT research systems
  • Novel and effective feedback collection mechanisms
  • Monitoring of user adoption of research systems and the level and quality of feedback
  • Improvements in the quality of feedback through better prompting and collection mechanisms
  • Growth and increase of feedback over time
  • Build distinct communities of users
  • Aggregation of user feedback data collections for project R&D


  • Develop methods for automatically annotating and analyzing translation requests, SMT system outputs, and user feedback in order to: understand what users submit for translation, guide the collection of user feedback, and understand how users judge translation quality.
  • Develop a robust set of linguistic metrics that are able to correlate well with human judgements according to several translation quality aspects and work for several languages of the project. Produce an open source suite of automatic metrics for SMT.
  • Produce linguistic annotation tools to be used throughout the project. This includes: 1) Develop novel parsing techniques for analyzing noisy translation output; 2) Develop domain adaptation methods so that a variety of linguistic analyzers can be run on parallel corpora from many different domains.


  • Create baseline MT systems for all language pairs in the project
  • MT system training and adaptation to user feedback
  • MT confidence for guiding user feedback
  • Search strategies for user-feedback-constrained re-translation
  • Robust Natural Language Generation (NLG) and its incorporation into fluent MT


  • Common and consistent use of resources
  • Consistent evaluation of all component technologies;
  • Internal evaluation of integrated systems, research and real-time demonstrators
  • Demonstrate that the quality of the research within the project is at the state-of-the art by participation in international evaluations of translation technology
  • Release Open Source modelling and analysis tools developed within the project
  • Distribute data collected within the project for use by other researchers
  • Establish as a direct dissemination mechanism for research results

Content: David Marecek.
Site is valid XHTML 1.0 and valid CSS.
2010 © Institute of Formal and Applied Linguistics. All Rights Reserved.

Site navigation: