Artificial intelligence-based approaches to reducing cheating in online education

Abstract

The rapid expansion of online education has increased concerns about academic integrity, particularly in remote examinations where traditional supervision is limited. This paper discusses the challenge of cheating in distance learning environments and explores the role of artificial intelligence in supporting reliable online assessment. By analyzing student interaction behaviors such as mouse movements, copy-and-paste activities, and time spent on exam questions, AI-based approaches can provide insights into behavioral patterns during assessments. The paper also highlights the potential use of machine learning models, including XGBoost, as a practical direction for future implementation, while emphasizing ethical considerations related to privacy and fairness.