Informative Manual Evaluation of Machine Translation Output
Maja Popović
Abstract
This work proposes a new method for manual evaluation of Machine Translation (MT) output based on marking actual issues in the translated text. The novelty is that the evaluators are not assigning any scores, nor classifying errors, but marking all problematic parts (words, phrases, sentences) of the translation.
Topics & Concepts
Computer scienceNoveltyMachine translationNatural language processingRank (graph theory)Artificial intelligenceTranslation (biology)Domain (mathematical analysis)Quality (philosophy)MathematicsPsychologyBiochemistrySocial psychologyPhilosophyGeneChemistryEpistemologyMessenger RNAMathematical analysisCombinatoricsNatural Language Processing TechniquesTopic ModelingSemantic Web and Ontologies