One Tree is not Enough: Cross-lingual Accumulative Structure Transfer for Semantic Indeterminacy
Patrick Ziering, Lonneke van der Plas · Recent Advances in Natural Language Processing · 2015
We address the task of parsing semantically indeterminate expressions, for which several correct structures exist that do not lead to differences in meaning. We present a novel non-deterministic structure transfer method that accumulates all structural information based on cross-lingual word distance derived from parallel corpora. Our system’s output is a ranked list of trees. To evaluate our system, we adopted common IR metrics. We show that our system outperforms previous cross-lingual structure transfer methods significantly. In addition, we illustrate that tree accumulation can be used to combine partial evidence across languages to form a single structure, thereby making use of sparse parallel data in an optimal way.