We have developed a two-stage machine translation system. The first stage is a rule-based machine translation system. The second stage is a statistical machine translation system. Our goal with this system was to obtain fewer unknown words and fewer ungrammatical sentences. The results that we obtained in this experiments were so good.
We did not use all training data nor optimize parameters nor did we use a reordering model. In future experiments, we will try these data and these techniques, which we expect will enable our system to perform better.