Post-MT Term Swapper: Supplementing a Statistical Machine Translation System with a User Dictionary

A statistical machine translation (SMT) system requires homogeneous training data in order to get domain-sensitive (or context-sensitive) terminology translations. If the data consists of various domains, it is difficult for an SMT system to learn context-sensitive terminology mappings probabilistically. Yet, terminology translation accuracy is an important issue for MT users. This paper explores an approach to tackle this terminology translation problem for an SMT system. We propose a way to identify terminology translations from MT output and automatically swap them with user-defined translations. Our approach is simple and can be applied to any type of MT system. We call our prototype “Term Swapper.” Term Swapper allows MT users to draw on their own dictionaries without affecting any parts of the MT output except for the terminology translation(s) in question. Using an SMT system developed at Microsoft Research, called MSR-MT (Quirk, et al., (2005); Menezes & Quirk (2005)), we conducted initial experiments to investigate the coverage rate of Term Swapper and its impact on the overall quality of MT output. The results from our experiments show high coverage and positive impact on the overall MT quality.