Özet:

This study evaluates the appearance, development and the point where machine translation comes today, referring to the machine translation methods that are widely used today through popular machine translation systems (Google Translation, Microsoft Bing Translation, Yandex Translation). Google Translate, which has been trained on the United Nations documents containing 20 billion words, in the early years used the method of "Rural-Based Machine Translation" and the method of "Statistical Machine Translation" and in recent years began to use the method of "Synirical Machine Translation", which is closer to human translation. This method is used in 9 languages, including English, French, German, Spanish, Portuguese, Chinese, Japanese, Korean and Turkish. The Yandex Translation, which offers translation services in more than 95 languages, does translation by analyzing the languages on which it works by statistical methods, not according to the pre-learned language rules. By developing this method in recent years, Yandex has begun to use a hybrid model (statistical machine translation and nervous machine translation together). Microsoft Bing Translate provides text translation services in more than 70 languages, 60 of which use the method of nervous machine translation and 10 in statistical machine translation. In the article, the statistical machine translation is assessed by examples of English-Turkish, while the nervous machine translation is assessed by examples of English-Turkish.

Anahtar Kelimeler: