Application of Recurrent Neural Networks for User Verification based on Keystroke Dynamics

• Autorzy: Kobojek P. , Saeed K.
• Języki publikacji: EN

Abstrakty

EN

Keystroke dynamics is one of the biometrics techniques that can be used for the verification of a human being. This work briefly introduces the history of biometrics and the state of the art in keystroke dynamics. Moreover, it presents an algorithm for human verification based on these data. In order to achieve that, authors’ training and test sets were prepared and a reference dataset was used. The described algorithm is a classifier based on recurrent neural networks (LSTMand GRU). High accuracy without false positive errors as well as high scalability in terms of user count were chosen as goals. Some attempts were made to mitigate natural problems of the algorithm (e.g. generating artificial data). Experiments were performed with different network architectures. Authors assumed that keystroke dynamics data have sequence nature, which influenced their choice of classifier. They have achieved satisfying results, especially when it comes to false positive free setting.

Słowa kluczowe

EN

biometrics; GRU networks; keystroke dynamics; LSTM networks; recurrent neural network; user verification

• Wydawca: Instytut Łączności - Państwowy Instytut Badawczy
• Czasopismo: Journal of Telecommunications and Information Technology
• Rocznik: 2016
• Tom: nr 3
• Strony: 80--90
• Opis fizyczny: Bibliogr. 30 poz., rys., tab.

Twórcy

autor

Kobojek P.

• Faculty of Mathematics and Information Sciences, Warsaw University of Technology Koszykowa st 75, 00-662 Warsaw, Poland

autor

Saeed K.

• Faculty of Mathematics and Information Sciences, Warsaw University of Technology Koszykowa st 75, 00-662 Warsaw, Poland

Bibliografia

• [1] R. Gaines, W. Lisowski, S. J. Press, and N. Shapiro, “Authentication by keystroke timing: some preliminary results”, R-2526-NSF RAND Report, RAND Corporation, Santa Monica, CA, USA, May 1980.
• [2] F. Monrose and A. D. Rubin, “Keystroke dynamics as a biometric for authentication”, Future Gener. Comp. Syst., vol. 16, pp. 351–359, 2000.
• [3] R. Joyce and G. Gupta, “Identity authorization based on keystroke latencies”, Commun. of the ACM, vol. 33, no. 2, pp. 168–176, 1990.
• [4] D. Mahar, R. Napier, M. Wagner, W. Laverty, R. Henderson, and M. Hiron, “Optimizing digraph-latency based biometric typist verification systems: Inter and intra typists differences in digraph latency distributions”, Int. J. Human-Comp. Stud., vol. 43, no. 4, pp. 579–592, 1995.
• [5] P. R. Dholi and K. P. Chaudhari, “Typing Pattern Recognition Using Keystroke Dynamics”, in Mobile Communication and Power Engineering, Vi. V. Das and Y. Chaba, Eds. Communications in Computer and Information Science, vol. 296, pp. 275–280. Springer, 2012.
• [6] G. Ho, “TapDynamics: Strengthening User Authentication on Mobile Phones with Keystroke Dynamics”, Tech. Rep., Stanford University, San Francisco, CA, USA, 2014.
• [7] Y. Deng and Y. Zhong, “Keystroke dynamics advances for mobile devices using deep neural network”, in Recent Advances in User Authentication Using Keystroke Dynamics Biometrics, Y. Zhong and Y. Deng, Eds. Science Gate Publishing, 2015, vol. 2, pp. 59–70.
• [8] C. Giuffrida, K. Majdanik, M. Conti, and H. Bos, “I sensed it was you: Authenticating mobile users with sensor-enhanced keystroke dynamics”, in Detection of Intrusions and Malware, and Vulnerability Assessment, S. Dietrich, Ed. LNCS, vol. 8550, pp. 92–111. Springer, 2014 (doi: 10.07/978/3-319-08509-8 6).
• [9] M. Rogowski, K. Saeed, M. Rybnik, M. Tabędzki, and M. Adamski, “User authentication for mobile devices”, in Computer Information Systems and Industrial Management, K. Saeed, R. Chaki, A. Cortesi, and S. Wierzchoń, Eds. LNCS, vol. 8104, pp. 47–58. Springer, 2013.
• [10] T. Bhattasali and K. Saeed, “Two Factor Remote Authentication in Healthcare”, in Proc. 3rd Int. Conf. Advan. in Comput., Commun. & Inform. ICACCI 2014, Delhi, India, 2014, pp. 380–386.
• [11] T. Bhattasali, K. Saeed, N. Chaki, and R. Chaki, “Bio-authentication for layered remote health monitor framework”, J. Medical Inform. & Technol., vol. 23, no. 1, pp. 131–139, 2014.
• [12] M. Rybnik, P. Panasiuk, K. Saeed, and M. Rogowski, “Advances in the keystroke dynamics: the practical impact of database quality”, in Computer Information Systems and Industrial Management, A. Cortesi, N. Chaki, K. Saeed, and S. Wierzchoń, Eds. LNCS, vol. 7564, pp. 203–214. Springer, 2012 (doi: 101007/978-3-64233260-9 17).
• [13] K. S. Killourhy and R. A. Maxion, “Comparing anomaly detectors for keystroke dynamics”, in Proc. 39th Annual IEEE/IFIP Int. Conf. Dependable Syst. & Netw. DSN 2009, Lisbon, Portugal, 2009, pp. 125–134.
• [14] Y. Deng and Y. Zhong, “Keystroke Dynamics User Authentication Based on Gaussian Mixture Model and Deep Belief Nets”, ISRN Sig. Process., vol. 2013, article ID 565183, 2013 (doi: 10.1155/2013/565183).
• [15] “ICT Facts and Figures” 2005, 2010, 2014, Telecommunication Development Bureau, International Telecommunication Union (ITU), 24 May 2015.
• [16] S. Hochreiter and J. Schmidhuber, “Long short-term memory”, Neural Computat., vol. 9, no. 8, pp. 1735–1780, 1997.
• [17] Y. Bengio, P. Simard, and P. Frasconi, “Learning long-term dependencies with gradient descent is difficult”, IEEE Trans. on Neural Netw., vol. 5, no. 2, pp. 157–166, 1994.
• [18] W. Zaremba, I. Sutskever, and O. Vinyals, “Recurrent Neural Network Regularization”, in Proc. Int. Conf. on Learn. Representat. ICLR 2015 San Diego, CA, USA, 2015 [Online]. Available: https://arxiv.org/pdf/1409.2329.pdf
• [19] S. K. Sønderby, C. K. Sønderby, H. Nielsen, and O. Winther, “Convolutional LSTM Networks for subcellular localization of proteins”, in Algorithms for Computational Biology, A.-H. Dediu, F. Hernández-Quiroz, C. Martín-Vide, and D. A. Rosenblueth, Eds. LNCS, vol. 9199, pp. 68–80. Springer, 2015.
• [20] C. Olah, “Understanding LSTM Networks”, Aug. 27, 2015 [Online]. Available: http://colah.github.io/posts/2015-08-UnderstandingLSTMs/ (accessed on Aug. 9, 2016)
• [21] K. Cho, F. Bougares, H. Schwenk, D. Bahdanau, and Y. Bengio, “Learning phrase representations using RNN Encoder-decoder for statistical machine translation”, in Proc. of the Conf. on Empir. Methods in Natural Language Process. EMNLP 2014, Doha, Qatar, 2014, pp. 1724–1734.
• [22] P. J. Werbos, “Backpropagation through time: What it does and how to do it”, Proc. of the IEEE, vol. 78, no. 10, pp. 1550–1560, 1998.
• [23] D. P. Kingma and J. L. Ba, “Adam: A Method for stochastic optimization”, in Proc. Int. Conf. on Learn. Representat. ICLR 2015 San Diego, CA, USA, 2015 [Online]. Available: https://arxiv.org/ pdf/1412.6980v8.pdf
• [24] A. Graves, M. Liwicki, S. Fernandez, R. Bertolami, H. Bunke, and J. Schmidhuber, “A novel connectionist system for unconstrained handwriting recognition”, IEEE Trans. Pattern Anal. & Mach. Intellig., vol. 31, no. 5, pp. 855–868, 2009.
• [25] A. Ng, “Feature selection, L1 vs. L2 regularization, and rotational invariance”, in Proc. 21st Int. Conf. on Machine Learn. ICML’04, Banff, Canada, 2004.
• [26] N. Srivastava et al., “Dropout: A simple way to prevent neural networks from overfitting”, J. of Mach. Learn. Res., vol. 15, no. 1, 1929–1958, 2014.
• [27] K. Killourhy and R. Maxion, “Keystroke Dynamics – Benchmark Data Set”, Carnegie Mellon University, Pittsburgh, PA, USA [Online]. Available: http://www.cs.cmu.edu/∼keystroke/ (accessed on May 27, 2016).
• [28] H. He and E. A. Garcia, “Learning from imbalanced data”, IEEE Trans. on Knowl. & Data Engin., vol. 21, no. 9, pp. 1263–1284, 2009.
• [29] J. Schneider, “Cross Validation”, Carnegie Mellon University, Pittsburgh, PA, USA [Online]. Available: https://www.cs.cmu.edu/∼schneide/tut5/node42.html (accessed on May 27, 2016).
• [30] P. Malhotra, L. Vig, G. Shroff, and P. Agarwal, “Long short term memory networks for anomaly detection in time series”, in Proc. of European Symp. on Artif. Neural Netw., Computat. Intellig. and Machine Learning ESANN 2015, Bruges, Belgium, 2015.