Modeling Inflection and Word-Formation in SMT
Alexander Fraser, Marion Weller, Aoife Cahill, Fabienne Cap
EACL 2012

The current state-of-the-art in statistical machine translation (SMT)
suffers from issues of sparsity and inadequate modeling power when
translating into morphologically rich languages.  We model both
inflection and word-formation for the task of translating into German.
For inflection, we generalize over different inflected forms of a
word, while also ensuring coherence of the inflected output.  For
word-formation, we address compounding, which is highly productive in
German. For both inflection and word-formation, we address the problem
of portmanteaus.  We translate from English words to an underspecified
German representation and then use linear-chain CRFs to predict the
fully specified German representation.  We show that improved modeling
of inflection and word-formation leads to improvement in translation
performance.