SHEF-NN: Translation Quality Estimation with Neural Networks

Kashif Shah, Varvara Logacheva, Gustavo Paetzold, Frédéric Blain, Daniel Beck, Fethi Bougares, Lucia Specia
2015 Proceedings of the Tenth Workshop on Statistical Machine Translation  
We describe our systems for Tasks 1 and 2 of the WMT15 Shared Task on Quality Estimation. Our submissions use (i) a continuous space language model to extract additional features for Task 1 (SHEF-GP, SHEF-SVM), (ii) a continuous bagof-words model to produce word embeddings as features for Task 2 (SHEF-W2V) and (iii) a combination of features produced by QuEst++ and a feature produced with word embedding models (SHEF-QuEst++). Our systems outperform the baseline as well as many other
more » ... The results are especially encouraging for Task 2, where our best performing system (SHEF-W2V) only uses features learned in an unsupervised fashion. Continuous Space Language Model Features for QE Neural networks model non-linear relationships between the input features and target outputs.
doi:10.18653/v1/w15-3041 dblp:conf/wmt/ShahLPBBBS15 fatcat:rykcrg6575acbaoa5hbm7jnsgy