Detecting AI-generated essays with lexical stylometric features using machine learning

Sumbula, Tahmida Haque and Sulaiman, Suriani and Hj. Ariffin, Adlina (2026) Detecting AI-generated essays with lexical stylometric features using machine learning. International Journal on Perceptive and Cognitive Computing (IJPCC), 12 (2). pp. 11-18. E-ISSN 2462-229X

PDF - Published Version Restricted to Registered users only Download (552kB) | Request a copy

Abstract

The increasing use of generative artificial intelligence (AI) in academic writing has intensified concerns about authorship authenticity. This study investigates whether lexical stylometric features alone can distinguish AI-generated essays from human-written essays. A balanced dataset of 200 undergraduate essays (100 AI-generated and 100 human-written) was analysed using vocabulary richness measures, lexical density, word usage distributions, and word-level n-grams. Logistic Regression, Decision Tree, and Support Vector Machine classifiers were applied to evaluate performance. The results indicated that lexical features provide strong discriminatory capability across the three models with accuracy ranging between 85 – 90% for AI-generated essays and 90-100% for human-written essays, suggesting that effective AI authorship detection can be achieved without complex feature engineering or advanced stylometric features. However, the findings should be interpreted as evidence for a lightweight and interpretable baseline rather than a fully generalizable detector, since robustness against unseen prompts, unseen AI models, paraphrasing, and broader writing domains were not tested. The study contributes to an interpretable lexical-feature framework for AI authorship detection in educational contexts and identifies the need for future validation using different categories of stylometric features, as well as larger, multi-modal, and cross-domain datasets

Item Type:	Article (Journal)
Uncontrolled Keywords:	machine learning, lexical features, stylometric, AI-generated essays, academic writing, ChatGPT
Subjects:	T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Islamic Revealed Knowledge and Human Sciences > Department of English Language & Literature Kulliyyah of Islamic Revealed Knowledge and Human Sciences Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology Kulliyyah of Information and Communication Technology > Department of Computer Science Kulliyyah of Information and Communication Technology > Department of Computer Science
Depositing User:	AP Dr Adlina Ariffin
Date Deposited:	10 Aug 2026 16:40
Last Update:	10 Aug 2026 16:40
Queue Number:	2026-08-Q4621
URI:	http://irep.iium.edu.my/id/eprint/130655

Actions (login required)

View Item