Paper: Reducing Annotation Effort for Quality Estimation via Active Learning

ACL ID P13-2097
Title Reducing Annotation Effort for Quality Estimation via Active Learning
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

Quality estimation models provide feed- back on the quality of machine translated texts. They are usually trained on human- annotated datasets, which are very costly due to its task-specific nature. We in- vestigate active learning techniques to re- duce the size of these datasets and thus annotation effort. Experiments on a num- ber of datasets show that with as little as 25% of the training instances it is possible to obtain similar or superior performance compared to that of the complete datasets. In other words, our active learning query strategies can not only reduce annotation effort but can also result in better quality predictors.