An Investigation on the Effectiveness of Features for Translation Quality Estimation
Kashif Ur Rehman Shah, Trevor Conn, Lucia Specia · 2013
We describe a systematic analysis on the effectiveness of features commonly ex-ploited for the problem of predicting ma-chine translation quality. Using a fea-ture selection technique based on Gaus-sian Processes, we identify small subsets of features that perform well across many datasets for different language pairs, text domains, machine translation systems and quality labels. In addition, we show the potential of the reduced feature sets result-ing from our feature selection technique to lead to significantly better performance in most datasets, as compared to the complete feature sets. 1