https://github.com/njanakiev/keystroke-biometrics

Biometrics on Keystroke Dynamics
https://github.com/njanakiev/keystroke-biometrics

biometrics classification data-science keras keystroke-dynamics python sklearn

Last synced: 3 months ago
JSON representation

Biometrics on Keystroke Dynamics

• Host: GitHub
• URL: https://github.com/njanakiev/keystroke-biometrics
• Owner: njanakiev
• License: mit
• Created: 2018-05-24T18:42:52.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2018-06-07T11:11:06.000Z (about 7 years ago)
• Last Synced: 2025-03-24T19:08:07.248Z (3 months ago)
• Topics: biometrics, classification, data-science, keras, keystroke-dynamics, python, sklearn
• Language: Jupyter Notebook
• Size: 14.7 MB
• Stars: 38
• Watchers: 1
• Forks: 11
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Biometric Prediction on Keystroke Dynamics

This project takes a look at a biometric classifier with keystroke dynamics. Besides typing a password to validate a user, it is possible to discriminate users by their typing rhythms. This features can be used to detect impostors. This project is based on the data set of the publication _Killourhy, K. S., & Maxion, R. A. [Comparing anomaly detectors for keystroke dynamics.](http://www.cs.cmu.edu/%7Emaxion/pubs/KillourhyMaxion09.pdf) In Proc. of the Int. Conf. on Dependable Systems and Networks (DSN) (pp. 125-134)_. The objective is to build a classification baser on neural networks with the [keras](https://keras.io/) library.

The [data set](https://www.cs.cmu.edu/~keystroke/) consists of 51 subject each typing the password `.tie5Roanl` 400 times. The measures collected are the hold time (H), keydown-keydown time (DD) and keyup-keydown time (UD).

![tsne](tsne.png)

# License 

This project is licensed under the MIT license. See the [LICENSE](LICENSE) for details