Automated Machine Translation Improvement Through Post-Editing Techniques: Analyst and Translator Experiments.
Jennifer B. Doyon, Christine Doran, C. Donald Means, D. Kermode Parr · 2008
Technology (NIST) (NIST, 2008) MT evalua-tions of today, the U.S. Government has been instrumental in establishing measurements and baselines for the state-of-the-art in MT en-gines. In the same vein, the Automated Ma-chine Translation Improvement Through Post-Editing Techniques (PEMT) project sought to establish a baseline of MT engines based on the perceptions of potential users. In contrast to these previous evaluations, the PEMT project’s experiments also determined the mi-nimal quality level output needed to achieve before users found the output acceptable. Based on these findings, the PEMT team in-vestigated using post-editing techniques to achieve this level. This paper will present ex-periments in which analysts and translators were asked to evaluate MT output processed with varying post-editing techniques. The re-sults show at what level the analysts and trans-lators find MT useful and are willing to work with it. We also establish a ranking of the types of post-edits necessary to elevate MT output to the minimal acceptance level. 1