Statistical Machine Translation from Slovenian to English

Abstract

In this paper, we analyse three statistical models for the machine translation of Slovenian into English. All of them are based on the IBM Model~4, but differ in the type of linguistic knowledge they use. Model 4a uses only basic linguistic units of the text, i.e., words and sentences. In Model 4b, lemmatisation is used as a preprocessing step of the translation task. Lemmatisation also makes it possible to add a Slovenian-English dictionary as an additional knowledge source. Model 4c takes advantage of the morpho-syntactic descriptions (MSD) of words. In Model 4c, MSD codes replace the automatic word classes used in Models 4a and 4b. The models are experimentally evaluated using the IJS-ELAN parallel corpus.