Identifying potential cheaters by tracking their behaviors through mouse activities

Published: 24 June 2020| Version 2 | DOI: 10.17632/fxp4n6b7h5.2


In this paper, the authors examine the developed mouse tracking application along with developed Moodle plugin in a blended course mid-term (20%) examination for the purpose of detecting and identifying the potential cheaters. The proposed model correctly predicted 94% of students committing illicit actions during the online mid-term examination, which can be possible to early intervene and prevent illegal actions. The study outcome can be used to analyze the learners’ mouse tracking behaviors that lead to a better process of secret reconstruction and transparent space.



Machine Learning Algorithm, Academic Assessment, Academic Intervention, Cheater Detection