Semi-supervised dependency parsing using generalized tri-training

Martins et al. (2008) presented what to the best of our knowledge still ranks as the best overall result on the CONLLX Shared Task datasets. The paper shows how triads of stacked dependency parsers described in Martins et al. (2008) can label unlabeled data for each other in a way similar to co-training and produce end parsers that are significantly better than any of the stacked input parsers. We evaluate our system on five datasets from the CONLL-X Shared Task and obtain 10

• Type: Journal
• Year: 2010
• Where: COLING
• Authors: Anders Søgaard, Christian Rishøj
• Keywords: COLING 2010, Computational Linguistics, CoNLLX Shared Task, Parsers, Stacked Input Parsers