It is a way that is being considered in research. It consists in breaking up the phrase in little parts called chunks. I didn't quite understand how the lack of context in this technique, was going to circumvent the context problem. I expected that some approach of improving MT's accuracy would be the way to go. Chunking is done to improve throughput. Along with it, there has to be others part of the translation task would need to be improved also, to keep the overall productivity. Interested in the subject of chunking I remain.
The contents of this post will automatically be included in the ticket generated. Please add any additional comments or explanation (optional)