Skip to content
Conference Open access

Human-in-the-Loop Machine Learning for Adaptive Communication Assessment in Humanities Education: A Systematic Survey and Research Framework

2026 · EPJ Web of Conferences · 0 citations · 19 references

Abstract

The fast development of digital education has raised the necessity for scalable, clear and pedagogically relevant systems that can assess communication skills in humanities. In contrast with the objective areas of study, humanities assessment is associated with interpretive reasoning, rhetorical development and subjectivity. Therefore, standard machine-aided assessment tools are not applicable. Moreover, currently available machine learning based educational platforms lack sensitivity to domain, ex-plainability and structured approach to cooperation with faculties, which prevents them from being used in humanities departments. To address these challenges this paper offers a review and introduces a research framework for machine-learning-based adaptive systems in humanities education. The framework highlights the importance of involving faculty expertise throughout the development and evaluation stages of these systems. It includes semantic modeling using transformers, multi-objective assessement, reinforcement learning for personalization, and explanation layer within continuous improvement cycle of faculties. The study synthesizes existing research, identifies key technical and ethical challenges, and outlines emerging directions for explainable and privacy-preserving educational AI. The findings suggest that collaborative, transparent, and adaptive ML systems can enhance formative feedback processes while preserving academic integrity in humanities communication assessment.

Read PDF

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.