Paper: Travatar: A Forest-to-String Machine Translation Engine based on Tree Transducers

ACL ID P13-4016
Title Travatar: A Forest-to-String Machine Translation Engine based on Tree Transducers
Venue Annual Meeting of the Association of Computational Linguistics
Session System Demonstration
Year 2013
Authors

In this paper we describe Travatar, a forest-to-string machine translation (MT) engine based on tree transducers. It pro- vides an open-source C++ implementation for the entire forest-to-string MT pipeline, including rule extraction, tuning, decod- ing, and evaluation. There are a number of options for model training, and tuning includes advanced options such as hyper- graph MERT, and training of sparse fea- tures through online learning. The train- ing pipeline is modeled after that of the popular Moses decoder, so users famil- iar with Moses should be able to get started quickly. We perform a valida- tion experiment of the decoder on English- Japanese machine translation, and find that it is possible to achieve greater accuracy than translation using phrase-based and hierarchical-phrase-...