Statistical language approach translates into success

Machine translation currently translates text from one language into another. However, work is under way on speech-to-speech translation. There are two kinds of machine translation: knowledge-based and statistical. Knowledge-based systems translate documents by converting words and grammar directly from one language into another. Rather than using the knowledge-based system's direct word-by-word translation techniques, statistical approaches translate documents by statistically analyzing entire phrases and, over time, "learning" how various languages work. The article examines the pros and cons of both systems, and predicts that statistical methods will become more popular, however the future will involve combining statistical and knowledge-based methods to create better systems.