Search and Filter

Early Prediction Of AI-Assisted Cheating Risk In Online Exams Through Learning Analytics

Date

AI Overview

Why it matters

Unauthorized generative AI use poses risks to exam integrity. Most detection methods monitor during exams or inspect submissions afterwards, offering limited opportunity to identify individual students who may benefit from early academic support.

Study at a glance
  • Question: Can students' early-semester learning management system activity predict their risk of AI-assisted cheating during a final exam?
  • Setting and data: A public university in Türkiye, 52 first-year undergraduates in an Introduction to Programming course, learning management system logs and face-to-face proctored exam logs.
  • Method: Machine learning classification using leave-one-out cross-validation and fold-specific feature selection.
Key findings
  • Logistic regression predicted AI-assisted cheating risk from first-eight-week learning management system traces with 73.1% accuracy, a balanced accuracy of 0.718, and an area under the curve of 0.763.
  • Course-module views, assignment submissions, and video-access days were the most consistently selected predictors across folds, with all showing lower activity levels in the high-risk group.
  • During the proctored final exam, 44.2% of the undergraduates were categorized as high-risk based on repeated suspicious actions such as copying, focus loss, and right-clicking.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

AI-assisted cheating has become an important threat to the security of online exams. This study examines whether the risk of AI-assisted cheating in the final exam can be predicted using students' digital traces in the learning management system (LMS) during the first eight weeks of the semester. The sample comprised 52 first-year undergraduates enrolled in a bachelor's program in Computer Education and Instructional Technology and taking an Introduction to Programming course at a public university in Turkiye. Students were labeled as low- or high-risk based on suspicious behaviors recorded in the final-exam logs, including copy, focus-loss, and right-click events. Of the 52 students, 23 (44.2%) were labeled as high-risk in a proctored, face-to-face exam. Group membership was then predicted using five features selected from 27 candidates extracted from students' digital traces. Logistic Regression, Naive Bayes, Random Forest, and Gradient Boosting algorithms were used to build the prediction models. Model performance was evaluated using leave-one-out cross-validation (LOOCV) with fold-specific preprocessing and feature selection. Logistic Regression achieved the best performance (Accuracy = 73.1%). The results indicate that LMS interaction data can provide an early signal of AI-assisted cheating risk. Course-module views, assignment submissions, and the number of days on which course videos were accessed were the most consistently selected features across the LOOCV folds. These predictions are intended to support timely academic guidance, not to establish misconduct or initiate disciplinary action.

Research Categories

What is the application?
Who is the user?
Who age?
Why use AI?

Publication Source

Publisher
arXiv