Researchers See ‘Promising Direction’ for Auto-Detection of Human vs Machine Translation
Researchers at the University of Groningen found that using training data from multiple source languages improved the accuracy of both monolingual and multilingual classifiers.