View in EDS PDF Full Text

Türkçeden Türk İşaret Diline makine çeviri sistemi.

Saved in:
Bibliographic Details
Title: Türkçeden Türk İşaret Diline makine çeviri sistemi.
Alternate Title: Machine translation system from Turkish to Turkish Sign Language.
Authors: Erten, Hüsnü1 husnu.erten@gazi.edu.tr, Arıcı, Nursal2 nursal@gazi.edu.tr
Source: Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi,. 2025, Vol. 40 Issue 3, p1715-1729. 15p.
Subjects: Sign language, Hard of hearing people, Turkish language, Machine translating, Translating & interpreting
Abstract (English): In this study, an automatic translation system from Turkish texts to Turkish Sign Language (TSL) has been proposed in order to contribute to the solution of the problems experienced by the hearing impaired in communication. The lack of parallel corpus in languages with few users is also valid for Turkish and TSL. This situation negatively affects the success of statistical-based translations from Turkish to TSL. With this study, a parallel data set of 3000 sentences from Turkish to Turkish Sign Language has been created and contributed to the elimination of the parallel data deficiency in the field. In the study, by using rule-based and statistical-based translation methods together, it has been tried to prevent the low level of translation success due to lack of data. The BLEU (Bilingual Evaluation Understudy) score obtained from the test of raw data has been 6.58 % and it has been increased to 10.58 % after morphological analysis. With the rule-based translation system made with the information obtained from the morphological analysis, the success of the statistical translation has been increased to the level of 32.92 %. In addition, for the first time, a temporal gap has been left between the word groups, and the misunderstandings caused by incorrect positioning have been tried to be reduced during the visualization of the translation in this study. [ABSTRACT FROM AUTHOR]
Abstract (Turkish): Bu çalışmada işitme engellilerin iletişimde yaşadığı problemlerin çözümüne katkı sağlamak amacıyla Türkçe metinlerden Türk İşaret Dili (TİD) arasında otomatik çeviri sistemi önerilmektedir. Az kullanıcısı olan dillerin genelinde yaşanan eşlenik veri eksikliği, Türkçe ve TİD için de geçerlidir. Bu durum Türkçeden TİD’e yapılan istatistiksel tabanlı çevirilerin başarısını olumsuz etkilemektedir. Bu çalışma ile Türkçeden Türk İşaret Diline 3000 cümlelik eşlenik veri seti oluşturularak alandaki eşlenik veri eksikliğinin giderilmesine katkı sağlanmıştır. Çalışmada kural tabanlı ve istatistiksel tabanlı çeviri yöntemleri birlikte kullanılarak, veri azlığından kaynaklı çeviri başarısı düşüklüğünün önüne geçilmeye çalışılmıştır. Ham verinin testinden elde edilen BLEU (Bilingual Evaluation Understudy) skoru, % 6,58 olup morfolojik analizden geçirilerek % 10,58 seviyesine çıkarılmıştır. Morfolojik analizden elde edilen bilgilerle yapılan kural tabanlı çeviri sistemi ile de istatistiksel çevirinin başarısı % 32,92 seviyesine çıkartılmıştır. Ayrıca, ilk defa bu çalışmada, çevirinin görselleştirilmesi aşamasında kelime grupları arasında zamansal boşluk bırakılarak yanlış konumlandırmadan kaynaklanan anlam yanılgıları azaltılmaya çalışılmıştır. [ABSTRACT FROM AUTHOR]
Copyright of Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, is the property of Gazi University, Faculty of Engineering & Architecture and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Links:
  – Type: pdflink
Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 187653785
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Türkçeden Türk İşaret Diline makine çeviri sistemi.
– Name: TitleAlt
  Label: Alternate Title
  Group: TiAlt
  Data: Machine translation system from Turkish to Turkish Sign Language.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Erten%2C+Hüsnü%22">Erten, Hüsnü</searchLink><relatesTo>1</relatesTo><i> husnu.erten@gazi.edu.tr</i><br /><searchLink fieldCode="AR" term="%22Arıcı%2C+Nursal%22">Arıcı, Nursal</searchLink><relatesTo>2</relatesTo><i> nursal@gazi.edu.tr</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Journal+of+the+Faculty+of+Engineering+%26+Architecture+of+Gazi+University+%2F+Gazi+Üniversitesi+Mühendislik+Mimarlık+Fakültesi+Dergisi%2C%22">Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi,</searchLink>. 2025, Vol. 40 Issue 3, p1715-1729. 15p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Sign+language%22">Sign language</searchLink><br /><searchLink fieldCode="DE" term="%22Hard+of+hearing+people%22">Hard of hearing people</searchLink><br /><searchLink fieldCode="DE" term="%22Turkish+language%22">Turkish language</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+translating%22">Machine translating</searchLink><br /><searchLink fieldCode="DE" term="%22Translating+%26+interpreting%22">Translating & interpreting</searchLink>
– Name: Abstract
  Label: Abstract (English)
  Group: Ab
  Data: In this study, an automatic translation system from Turkish texts to Turkish Sign Language (TSL) has been proposed in order to contribute to the solution of the problems experienced by the hearing impaired in communication. The lack of parallel corpus in languages with few users is also valid for Turkish and TSL. This situation negatively affects the success of statistical-based translations from Turkish to TSL. With this study, a parallel data set of 3000 sentences from Turkish to Turkish Sign Language has been created and contributed to the elimination of the parallel data deficiency in the field. In the study, by using rule-based and statistical-based translation methods together, it has been tried to prevent the low level of translation success due to lack of data. The BLEU (Bilingual Evaluation Understudy) score obtained from the test of raw data has been 6.58 % and it has been increased to 10.58 % after morphological analysis. With the rule-based translation system made with the information obtained from the morphological analysis, the success of the statistical translation has been increased to the level of 32.92 %. In addition, for the first time, a temporal gap has been left between the word groups, and the misunderstandings caused by incorrect positioning have been tried to be reduced during the visualization of the translation in this study. [ABSTRACT FROM AUTHOR]
– Name: Abstract
  Label: Abstract (Turkish)
  Group: Ab
  Data: Bu çalışmada işitme engellilerin iletişimde yaşadığı problemlerin çözümüne katkı sağlamak amacıyla Türkçe metinlerden Türk İşaret Dili (TİD) arasında otomatik çeviri sistemi önerilmektedir. Az kullanıcısı olan dillerin genelinde yaşanan eşlenik veri eksikliği, Türkçe ve TİD için de geçerlidir. Bu durum Türkçeden TİD’e yapılan istatistiksel tabanlı çevirilerin başarısını olumsuz etkilemektedir. Bu çalışma ile Türkçeden Türk İşaret Diline 3000 cümlelik eşlenik veri seti oluşturularak alandaki eşlenik veri eksikliğinin giderilmesine katkı sağlanmıştır. Çalışmada kural tabanlı ve istatistiksel tabanlı çeviri yöntemleri birlikte kullanılarak, veri azlığından kaynaklı çeviri başarısı düşüklüğünün önüne geçilmeye çalışılmıştır. Ham verinin testinden elde edilen BLEU (Bilingual Evaluation Understudy) skoru, % 6,58 olup morfolojik analizden geçirilerek % 10,58 seviyesine çıkarılmıştır. Morfolojik analizden elde edilen bilgilerle yapılan kural tabanlı çeviri sistemi ile de istatistiksel çevirinin başarısı % 32,92 seviyesine çıkartılmıştır. Ayrıca, ilk defa bu çalışmada, çevirinin görselleştirilmesi aşamasında kelime grupları arasında zamansal boşluk bırakılarak yanlış konumlandırmadan kaynaklanan anlam yanılgıları azaltılmaya çalışılmıştır. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, is the property of Gazi University, Faculty of Engineering & Architecture and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=187653785
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.17341/gazimmfd.1283219
    Languages:
      – Code: tur
        Text: Turkish
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 1715
    Subjects:
      – SubjectFull: Sign language
        Type: general
      – SubjectFull: Hard of hearing people
        Type: general
      – SubjectFull: Turkish language
        Type: general
      – SubjectFull: Machine translating
        Type: general
      – SubjectFull: Translating & interpreting
        Type: general
    Titles:
      – TitleFull: Türkçeden Türk İşaret Diline makine çeviri sistemi.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Erten, Hüsnü
      – PersonEntity:
          Name:
            NameFull: Arıcı, Nursal
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 07
              Text: 2025
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 13001884
          Numbering:
            – Type: volume
              Value: 40
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Journal of the Faculty of Engineering & Architecture of Gazi University / Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi,
              Type: main
ResultId 1