Prolog Statistical Machine Translation is a fairly unsophisticated statistical machine translation program. It consists of a language model learner (which takes example sentences in the target language and learns a language model based on trigrams), a dictionary learner (which learns word-for-word translations), and a search program (which uses the data from the first two parts to translate a source sentence into the target language).
