• Corpus ID: 9355130

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

  title={Application of Recurrent Neural Networks for User Verification based on Keystroke Dynamics},
  author={Pawel Kobojek and Kashif Saeed},
  journal={Journal of telecommunications and information technology},
  • P. KobojekK. Saeed
  • Published 2016
  • Computer Science
  • Journal of telecommunications and information technology
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 (LSTM and GRU). High… 

User Keystroke Authentication and Recognition of Emotions Based on Convolutional Neural Network

Experimental studies showed that the proposed coding procedure made it possible to use CNN for analyzing Keystroke Dynamics and achieve recognition error of emotions and personality at the level of the best modern recognition systems.

Deep Neural Networks Applied to User Recognition Based on Keystroke Dynamics: Learning from Raw Data

The objective of this work is to demonstrate the hypothesis that user recognition algorithms applied to keystroke dynamics raw ( original) data can perform better than the ones that depend on Feature Engineering, and proposes a deep neural network named DRK.

Biometric Identification Based on Keystroke Dynamics

Various neural network layers—convolutional, recurrent, and dense—in different configurations were employed together with pooling and dropout layers and the results varied, with the best-achieved accuracy equal to 82% for the identification task.

Significant Parameters of the Keystroke for the Formation of the Input Field of a Convolutional Neural Network

The article shows that improvement of personality and emotion recognizers based on the keystroke pattern can be implemented using neural network solutions and justified the need for further research on improvement of keystroke parameters preprocessing procedures in order to increase their informational value.

Typing Biometrics for User Authentication-a One-shot Approach

A novel method is proposed, similar to that of FaceNet, where metric learning is used to learn an embedding of the high dimensional typing samples into a lower dimensional space, in which the task of performing comparisons between samples is greatly simplified.

Keystroke Dynamics using Auto Encoders

This paper will look at an Autoencoder model for learning the keystroke dynamics of specific users, which shows an improvement over the traditional tried and tested statistical approaches with an Equal Error Rate of 6.51%, with the additional benefits of relatively low training times and less reliance on feature engineering.

Free-Text Keystroke Dynamics for User Authentication

A novel feature engineering method is employed that generates image-like transition matrices for verifying user identity based on keystroke dynamics obtained from free-text and a convolution neural network with cutout achieves the best results.


This research focuses on a continuous authentication based on behavioral biometrics performed once the user has achieved control over the device and considers application usage patterns to evaluate and identify the current user, who may or may not be legitimate.


This research presents the design and implementation of a stand alone web based biometric keystroke authentication framework that creates a user’s keystroke typing profile and use it as a second form of authentication.



Keystroke Dynamics Advances for Mobile Devices Using Deep Neural Network

This work presents a deep learning approach to keystroke authentication, and shows how it can leverage a unique set of advanced build in sensors, including accelerometer and gyroscope to capture rich typing information than raw timing pattern.

Comparing anomaly-detection algorithms for keystroke dynamics

The objective is to collect a keystroke-dynamics data set, to develop a repeatable evaluation procedure, and to measure the performance of a range of detectors so that the results can be compared soundly.

Keystroke Dynamics User Authentication Based on Gaussian Mixture Model and Deep Belief Nets

The Gaussian mixture model with the universal background model (GMM-UBM) and the deep belief nets (DBN) are introduced, which are two generative model-based approaches to user authentication using keystroke dynamics that make no assumption about the underlying probability distribution and are fast for training and testing.

TapDynamics : Strengthening User Authentication on Mobile Phones with Keystroke Dynamics

This project construct and analyze four keystroke dynamic classifiers that use a smartphone’s sensors to learn the key tap behavior of the true owner and can be used to determine whether a login attempt is being made by thetrue owner or by an attacker who has guessed the owner's PIN.

Typing Pattern Recognition Using Keystroke Dynamics

A k-nearest neighbor approach is introduced to classify users’ keystroke dynamics profiles to authenticate user based on their typing samples.

Optimizing digraph-latency based biometric typist verification systems: inter and intra typist differences in digraph latency distributions

It was found that typist verification accuracy improved when a digraph-specific index of the distance between test and reference digraph latencies was employed and the utility of nonlinear modelling as a tool to establish optimum verification parameter settings was shown.

Keystroke dynamics as a biometric for authentication

Long Short Term Memory Networks for Anomaly Detection in Time Series

The efficacy of stacked LSTM networks for anomaly/fault detection in time series on ECG, space shuttle, power demand, and multi-sensor engine dataset is demonstrated.

Authentication by Keystroke Timing: Some Preliminary Results

The authors found that of the large number of digraphs represented in most ordinary paragraphs, there were five which could serve as a basis for distinguishing among the subjects, and that this method of distinguishing subjects might provide the basis for a computer authentication system.

Adam: A Method for Stochastic Optimization

This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.