2
Learning Bilingual Word Mappings
In this chapter and the next we examine how much can be accomplished in MT, working just at the level of words. We build up the framework for word-level alignment, leading to IBM models of statistical machine translation (SMT) in the next chapter. Learning bilingual word mappings from parallel corpora is the focus of this chapter.
In the previous chapter we saw that translation is a mapping or transformational process. Since the 1990s, borrowing techniques from speech, initiated by the IBM group, statistical machine translation slowly came to dominate MT. The basic idea in SMT is to teach a machine how to translate, through a large number of examples of translation. These examples should contain at least ...
Get Machine Translation now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
