Automatic error classification with multiple error labels
Although automatic classification of machine translation errors still cannot provide the same detailed granularity as manual error classification, it is an important task which enables estimation of translation errors and better understanding of the analyzed MT system, in a short time and on a large scale. State-of-the-art methods use hard decisions to assign single error labels to each word. This work presents first results of a new error classification method, which assigns multiple error labels to each word. We assign fractional counts for each label, which can be interpreted as a confidence for the label. Our method generates sensible multi-error suggestions, and improves the correlation between manual and automatic error distributions.