engleski

From Off-line Evaluation to On-line Selection

In this chapter we describe some of the problems that arise from the need to integrate four alternative translations for each input utterance, and to come up with exactly one optimal translation. In the center of this chapter is a learning method that was tailored to overcome the problem of incomparable confidence values delivered by the four competing translation paths. By using offline human feedback and applying a linear optimization heuristic, we determine a rescaling scheme that enable us to compare confidence values across modules. We also describe some additional information sources that further elaborate the selection procedure, and finally, outline some Quality of Service parameters that are supported by the selection module.

machine translation, evaluation, Verbmobil

nije evidentirano