Linguistic Features for Quality Estimation
Mariano Felice, Lucia Specia · 2012
This paper describes a study on the contribution of linguistically-informed features to the task of quality estimation for machine translation at sentence level. A standard regression algorithm is used to build models using a combination of linguistic and non-linguistic features extracted from the input text and its machine translation. Experiments with English-Spanish translations show that linguistic features, although informative on their own, are not yet able to outperform shallower features based on statistics from the input text, its translation and additional corpora. However, further analysis suggests that linguistic information is actually useful but needs to be carefully combined with other features in order to produce better results. 1