Open anoopsarkar opened 10 years ago
compare the unlabeled attachment accuracy against another implementation trained on the same training data and evaluated on the same dev data.
It seems that people are comparing to the Turbo dependency parser recently along with MSTParser or MALTparser.
See this paper from ACL 2014: http://people.csail.mit.edu/regina/my_papers/tens14.pdf
compare the unlabeled attachment accuracy against another implementation trained on the same training data and evaluated on the same dev data.