<![CDATA[The advancement of generative AI increases challenges in distinguishing human-written from AI-generated text, threatening academic integrity. This study aims to develop an AI-generated text detection model using a stylometric approach and machine learning algorithms. The stylometric approach was employed to extract 31 linguistic features representing lexical, syntactic, and readability characteristics of textual documents as the basis for classification using machine learning. Five machine learning algorithms, namely Naïve Bayes, Logistic Regression, Support Vector Machine (Linear and RBF), and Random Forest, were evaluated to identify the best-performing model. The experimental results indicate that the Random Forest model achieved the highest classification performance, with an accuracy of 97.94%. Feature importance analysis revealed that average word length was the most influential feature for distinguishing human-written text from AI-generated text. The best-performing model was subsequently implemented in a web-based application to support document originality verification. Overall, the findings demonstrate that combining stylometric analysis with machine learning offers a promising approach for detecting AI-generated text and supporting the preservation of academic integrity in the era of generative artificial intelligence.]]>
Copyrights © 2026
