Weakly supervised approaches for quality estimation

Erwan Moreau (1) , Carl Vogel (1)

1 School of Computer Science and Statistics [Dublin]

Currently, quality estimation (QE) is mostly addressed using supervised learning approaches. In this paper we show that unsupervised and weakly supervised approaches (using a small training set) perform almost as well as supervised ones, for a significantly lower cost. More generally, we study the various possible definitions, parameters, evaluation methods and approaches for QE, in order to show that there are multiple possible configurations for this task.

Mots clés: Quality estimation; Evaluation; Unsupervised learning; Weakly supervised learning

Domaines: Traitement du texte et du document

Erwan Moreau, Carl Vogel. Weakly supervised approaches for quality estimation. Machine Translation, 2013, 27 (3), pp 257-280. ⟨10.1007/s10590-013-9142-8⟩. ⟨hal-00860680⟩