Date of Publication :8th August 2024

Abstract:As a result of the widespread spread of education and assessment in the internet system, ensuring the integrity of online testing has become a major problem for educational institutions. Traditional proctoring methods may not be possible or scalable in an online environment, requiring the use of automated approaches. Machine learning offers promising opportunities to track students ' actions during online testing, allowing them to identify suspicious behavior that indicates academic dishonesty. This article provides a comprehensive overview of the process of monitoring the actions of students during online testing using machine learning methods. The process includes several basic steps: data collection, feature acquisition, data preprocessing, sample selection, training, evaluation, deployment, and feedback loop. Data collection involves various aspects of student interaction during online tests, including timestamps, mouse movement, keystrokes, and browser interaction. The research work was considered by students of the educational programs "6B01511-Informatics", "7M01511-Informatics", "7M01525-STEM education", "8D01511-Informatics" of the Eurasian National University named after L. N. Gumilyov. As a result, using machine learning to monitor students ' actions during online testing is a promising way to improve the integrity and security of online assessments while reducing the burden on teachers and administrators.

Reference :