Deciphering Manuscript Texts with the Assistance of Artificial Intelligence in Digitization Studies The Case of the eScriptorium Program Dijitalleştirme Çalışmalarında El Yazması Metinlerin Yapay Zekâ Yardımıyla Deşifre Edilmesi eScriptorium Programı Örneği


Creative Commons License

Karayiğit R.

Islam Arastirmalari Dergisi, vol.2025, no.53, pp.189-197, 2025 (Scopus, TRDizin) identifier

  • Publication Type: Article / Article
  • Volume: 2025 Issue: 53
  • Publication Date: 2025
  • Doi Number: 10.26570/isad.1530174
  • Journal Name: Islam Arastirmalari Dergisi
  • Journal Indexes: Scopus, ATLA Religion Database, Index Islamicus, MLA - Modern Language Association Database, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.189-197
  • Keywords: Artificial Intelligence, Digital Humanities, Digital Theological Studies, manuscripts
  • Open Archive Collection: AVESIS Open Access Collection
  • Istanbul University Affiliated: Yes

Abstract

In recent years, the growing interest in digital humanities within the field of theology has led researchers and academics to organize meetings on this subject. A series of conferences on Digital Theological Studies aims to inform researchers and guide them towards new interdisciplinary areas of study. Artificial intelligence models developed especially on Arabic texts contribute to more accurate reading of manuscripts. This article focuses on the practical use of the eScriptorium program for the digitization of manuscripts. Its purpose is to assess the advantages and disadvantages of the program, while also providing recommendations for solving the problems encountered in the scanned texts. In line with this objective example from within the program are presented to demonstrate how researchers should utilize it. Although eScriptorium is a program developed for the digitization of manuscripts through machine learning, it still requires manual intervention at certain levels. Therefore, it is necessary for engineers and social scientists to form collaborative working groups within the digital humanities field to improve existing programs and develop new ones. A deeper understanding of the logic of these programs and the functioning of artificial intelligence will contribute to the reduction of potential software errors over time. Consequently, the likelihood of obtaining near-accurate data through proper data entries will be strengthened throughout the process.