Mapping the Landscape of AI-Assisted L2 Writing Assessment: A Bibliometric and Trend-Forecasting Study (2021–2025)

Landscape of AI-Assisted L2 Writing Assessment

Authors

  • Nurgül Bekdemir Ordu University, Ordu, Turkey
  • Kadir Kesgin Bandirma Onyedi Eylul University, Balikesir, Bandirma, Turkey

Abstract

Language assessment has been undergoing a significant transformation with the rise of Artificial Intelligence (AI), particularly in writing, which is widely recognized as the most complex of the four language skills due to its multidimensional nature. Recent studies highlight how AI tools—such as ChatGPT—support feedback generation, streamline editing, and reduce teachers’ assessment workload. Despite the increasing volume of publications in AI-based writing assessment, there remains a lack of bibliometric studies that map the scholarly landscape of this emerging field. This study addresses that gap by conducting a bibliometric and predictive analysis of AI-based L2 writing assessment research published between 2021 and 2025. Data were retrieved from ScienceDirect, Wiley Online Library, SpringerLink, and SAGE Journals and analyzed using the BiBLoX platform for trend forecasting, topic modeling (LDA), and co-authorship network mapping. Citation predictions were modeled using machine learning algorithms, including Random Forest Regression. The results reveal evolving thematic focuses, leading contributors, and publication trends, offering a data-driven overview of the field and highlighting directions for future research.

Author Biography

Kadir Kesgin, Bandirma Onyedi Eylul University, Balikesir, Bandirma, Turkey

Kadir Kesgin is currently affiliated with the Department of Computer Technology at Bandirma Onyedi Eylul University, located in Bandirma, Balikesir, Turkey. 

Downloads

Published

2025-09-01