Formalization and Evaluation of Candidate Translation Generation
Estelle Maryline Delpech · 2014
This chapter presents the implementation of the translation generation method. The aim of this method is to translate various types of morphologically complex words and generate fertile translations. It starts by detailing the translation generation algorithm, and the evaluation of the results of the morphological splitting module. The chapter evaluates the quality of the generated translations. It describe the various experiments whose goal was to highlight the influence of data and generation strategies. Finally, the chapter discusses the limits of the work as well as the research prospects. To evaluate the translations generated, chapter uses the two references: a posteriori reference and a priori reference.