An enhanced English Arabic statistical machine translation approach using language knowledge / Mohammed Hassanien Mohammed Yousef ; Supervised Mervat Hassan Gheith , Tarek Elgazaly
Material type:
- تحسين النهج الاحصائى للترجمة الآلية من الانجليزية الى العربية باستخدام المعرفة اللغوية [Added title page title]
- Issued also as CD
Item type | Current library | Home library | Call number | Copy number | Status | Barcode | |
---|---|---|---|---|---|---|---|
![]() |
قاعة الرسائل الجامعية - الدور الاول | المكتبة المركزبة الجديدة - جامعة القاهرة | Cai01.18.07.M.Sc.2017.Mo.E (Browse shelf(Opens below)) | Not for loan | 01010110073964000 | ||
![]() |
مخـــزن الرســائل الجـــامعية - البدروم | المكتبة المركزبة الجديدة - جامعة القاهرة | Cai01.18.07.M.Sc.2017.Mo.E (Browse shelf(Opens below)) | 73964.CD | Not for loan | 01020110073964000 |
Thesis (M.Sc.) - Cairo University - Institute of Statistical Studies and Research - Department of Information Systems and Technology
Arabic is the language of the holy quran, and one of the six official languages of the UN. Although of being quite rich, Arabic language translation is very difficult, and Arabic machines translation are quite complex. In this thesis, an enhanced English Arabic statistical machine translation approach using language knowledge is introduced. Translation quality is an important measure for the translation system. In this research, translation quality is improved by a new proposed framework where language structural transfer knowledge is added to the traditional phrase based system (Moses). The new framework improves the quality of the translation system by about 1 BLUE for a large test set (451000 words), and about 2.5 BLUE for small test set (14326 words), without increasing the amount of training data, or radically changing the algorithms that can affect the translation engin
Issued also as CD
There are no comments on this title.