A Model Lexicalized Hierarchical Reordering for Phrase Based Translation
Vinh Van Nguyen, Thai Phuong Nguyen, Le-Minh Nguyen, Akira Shimazu · Procedia - Social and Behavioral Sciences · 2011
In this paper, we present a reordering model based on Maximum Entropy with local and non-local features. This model is extended from a hierarchical reordering model with PBSMT [1], which integrates rich syntactic information directly in decoder as local and non-local features of Maximum Entropy model. The advantages of this model are (1) maintaining the strength of phrase based approach with a hierarchical reordering model, (2) many kinds of rich linguistic information integrated in PBSMT as local and non-local features of MaxEntropy model. The experiment results with English-Vietnamese pair showed that our approach achieves significant improvements over the system which uses a lexical hierarchical reordering model [1].