Types of machine translation in academic text writing

Volume 1 , Issue 1 , September 2025 , Pages 89-110

Authors

آ.م.د. محمد حسین احمد 1

1 قسم اللغة الكردية - كلية التربية الأساسية - جامعة كرميان.

DOI logo 10.17656/jzsb.11711

Keywords

Abstract

The incorporation of the Kurdish language into Machine Translation neural networks constituted a significant milestone, particularly in the contemporary era wherein artificial intelligence assumes a pivotal role in advancing translational competencies and enhancing precision, processing velocity, and information reliability. Notwithstanding these technological advancements, the translation process continues to encounter substantive challenges and inherent limitations, predominantly attributable to linguistic diversity, grammatical complexities, and cultural specificities. This study, which is entitled “The Quality of Machine Translation in Academic Text Writing”, The research sample consists of fifty-nine (59) English sentences subjected to translation into Kurdish through the deployment of both Microsoft Translate and Google Translate systems, with the objective of conducting a comparative performance analysis between the two platforms. Following comprehensive analysis of the sample data and systematic identification of grammatical errors, the investigation culminated in the finding that both Google Translate and Microsoft Translate exhibit weak performance levels, with their capabilities demonstrating considerable similarity. This represents one of the most significant objectives and recommendations of the research, which we direct toward students and users, stipulating that these two systems should not be employed in the translation of scientific and academic texts. However, should their use become necessary under compelling circumstances, it is imperative that such utilization be restricted and that several conditions and guidelines be rigorously observed.

Statistics
  • Article view674
  • Downloads22
  • First online28 September 2025
  • Published at28 September 2025

  • RIS
  • BibTeX
  • EndNote
  • Mendeley
  • APA (7th edition)
  • MLA (9th edition)
  • Chicago
  • Harvard
  • IEEE
  • Vancouver