ASSESSMENT OF INSTAGRAM TRANSLATION OF NEWS FEEDS

dc.contributor.authorMahmoud Rashead Abuthaher
dc.date.accessioned2025-04-23T06:18:57Z
dc.date.available2025-04-23T06:18:57Z
dc.date.issued2023-04-09
dc.description.abstractThe present study assesses the Instagram’s automated English translation by concentrating on the translation accuracy of Arabic syntactic patterns and various types of metaphors. The research study focuses on textual coherence of selected Instagram’s informative, expressive, and operative texts, and measures levels of equivalence of texts containing the different types of metaphors between the source text and the target text. It is observed that the Instagram translation model errs at the level of syntactic patterns related to agents reversing and Arabic diacritics. It is also found that whenever the dataset lacks sufficient bilingual corpora, the machine encounters serious drawbacks in the translation. However, cases of translation prove that the Instagram translation model successfully produces communicative translations whenever it relies on rich parallel texts in the dataset. Self-attention by the Instagram model is also noted where it improves cases of translation in the second time of testing the translation quality but does not generate further enhancements even if the translation outcome still incurs serious translation issues. These two aspects will explore insights into future translation studies on machine translation.
dc.identifier.urihttps://hdl.handle.net/20.500.11888/20038
dc.language.isoen
dc.publisherجامعة النجاح الوطنية
dc.supervisorProf. Abdel Karim Daraghmeh
dc.titleASSESSMENT OF INSTAGRAM TRANSLATION OF NEWS FEEDS
dc.title.alternativeدراسة تقيمية لترجمة منشورات الانستغرام
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
محمود رشيد محمد ابو ظاهر.pdf
Size:
1.01 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Collections