Quality Estimation of Machine Translated Texts based on Direct Evidence from Training Data
This addresses the issue of meaning errors in fluent MT outputs for users relying on automated translation, though it appears incremental as it builds on existing quality estimation approaches.
The paper tackles the problem of estimating the quality of machine-translated texts without reference translations by using direct evidence from the training data of the MT system, showing that this simple method holds promise for any data-driven MT system.
Current Machine Translation systems achieve very good results on a growing variety of language pairs and data sets. However, it is now well known that they produce fluent translation outputs that often can contain important meaning errors. Quality Estimation task deals with the estimation of quality of translations produced by a Machine Translation system without depending on Reference Translations. A number of approaches have been suggested over the years. In this paper we show that the parallel corpus used as training data for training the MT system holds direct clues for estimating the quality of translations produced by the MT system. Our experiments show that this simple and direct method holds promise for quality estimation of translations produced by any purely data driven machine translation system.