Smart Evaluation of Stylistic Elements in Student Writing Using AI Tools

The integration of artificial intelligence in educational assessment has opened new pathways for evaluating nuanced aspects of student writing, such as stylistic and tonal features. This study proposes a novel AI-driven framework for analyzing stylistic elements in student literary compositions to enhance interpretive feedback and writing development. Existing methods for assessing literary essays often rely on rubric-based or manual evaluations, which are time-consuming, subjective, and limited in capturing subtle stylistic variations. To address these limitations, we introduce the Analyzing Tone using Explainable AI (AT-XAI) framework, which incorporates Stylometric Feature Extraction to analyze student essays automatically. This system evaluates tone, style, and linguistic patterns using interpretable AI models, offering transparent insights for educators and learners. The proposed method facilitates automated, consistent, and pedagogically relevant feedback, helping students refine their literary voice and critical writing. Findings show that AT-XAI achieves high accuracy in tone classification and stylistic feature detection, aligning closely with expert human evaluations while significantly reducing assessment time. The study demonstrates the potential of explainable AI to foster more equitable and scalable writing instruction across literature classrooms.

Перевод пока недоступен