Continuous authentication of students using behavioral biometrics in Online Judge environment

• Ronem Matos Lavareda Filho Universidade Federal do Amazonas
• Juan Gabriel Colonna Universidade Federal do Amazonas
• David Braga Fernandes Oliveira Federal University of Amazonas

DOI: https://doi.org/10.5753/cbie.sbie.2020.1193

Abstract

Student authentication in online judge environments is only verified at the beginning of the login session by password validation. However, in the case of exercises and assessments made in environments like this, there is a need for continuous and non-intrusive verification of the genuineness of students throughout the session and not just at login. In this article, we present a method to continuously authenticate students in online judge systems using behavioral biometrics, using specifically the coding dynamics. For this, a Siamese convolutional neural network architecture was designed that seeks to learn automatically, from a limited amount of raw data of the student's typing dynamics. In the experiment, our model achieves a recognition accuracy of 97.2%.

Keywords: Online judge, students, continuous authentication, behavioral biometrics

References

Chong, P., Elovici, Y., and Binder, A. (2019). User authentication based on mouse dynamics using deep neural networks: A comprehensive study. IEEE Transactions onInformation Forensics and Security.


Cruz, M. A. S., Duarte, J. C., and Goldschmidt, R. R. (2017). Dinâmica da digitação aplicada a autenticação periódica de usuários em ambientes virtuais de aprendizagem. Revista Brasileira de Informática na Educação, 25(02):36.


Galvão, L., Fernandes, D., and Gadelha, B. (2016). Juiz online como ferramenta de apoio a uma metodologia de ensino híbrido em programação. In Brazilian Symposium on Computers in Education (SBIE), volume 27, page 140.


Giot, R. and Rocha, A. (2019). Siamese networks for static keystroke dynamics authentication. In IEEE International Workshop on Information Forensics and Security.


Longi, K., Leinonen, J., Nygren, H., Salmi, J., Klami, A., and Vihavainen, A. (2015). Identification of programmers from typing patterns. InProceedings of the 15th KoliCalling Conference on Computing Education Research, pages 60–67. ACM.


Mondal, S. and Bours, P. (2017). A study on continuous authentication using a combina-tion of keystroke and mouse biometrics.Neurocomputing, 230:1–22.


Pisani, P. H. and Lorena, A. C. (2013). A systematic review on keystroke dynamics.Journal of the Brazilian Computer Society, 19(4):573–587.


Santos, V. A. et al. (2018). Siamesevo-depth: odometria visual através de redes neurais convolucionais siamesas.


Sekhar, C., Mukherjee, P., Guru, D. S., and Pulabaigari, V. (2019). Osvnet: Convolutional siamese network for writer independent online signature verification.arXiv preprintarXiv:1904.00240.


Ullah, A., Xiao, H., and Barker, T. (2019). A dynamic profile questions approach to mitigate impersonation in online examinations. Journal of Grid Computing, 17(2):209–223.


Xiaofeng, L., Shengfei, Z., and Shengwei, Y. (2019). Continuous authentication by free-text keystroke based on cnn plus rnn.Procedia computer science, 147:314–318.


Young, J. R., Davies, R. S., Jenkins, J. L., and Pfleger, I. (2019). Keystroke dynamics: establishing keyprints to verify users in online courses. Computers in the Schools,36(1):48–68.