Sequential Keystroke Behavioral Biometrics for Mobile User Identification via Multi-view Deep Learning

  title={Sequential Keystroke Behavioral Biometrics for Mobile User Identification via Multi-view Deep Learning},
  author={Lichao Sun and Yuqi Wang and Bokai Cao and Philip S. Yu and Witawas Srisa-an and Alex D. Leow},
With the rapid growth in smartphone usage, more organizations begin to focus on providing better services for mobile users. User identification can help these organizations to identify their customers and then cater services that have been customized for them. Currently, the use of cookies is the most common form to identify users. However, cookies are not easily transportable (e.g., when a user uses a different login account, cookies do not follow the user). This limitation motivates the need… 

Deepkey: Keystroke Dynamics and CNN for Biometric Recognition on Mobile Devices

Deep learning techniques, based on convolutional neural networks, are exploited to derive discriminative characteristics from the typing patterns of subjects entering personal identification numbers (PINs) and this paper studies the effectiveness of employing keystroke dynamics to discriminate between genuine and impostor users of mobile devices.

Keystroke Dynamics for Biometric Recognition in Handheld Devices

This paper evaluates the feasibility of performing keystroke-dynamics-based biometric verification by modeling the acquired sequences of typed characters with Gaussian mixture models.

To augment or not to augment? Data augmentation in user identification based on motion sensors

Experiments are conducted on a benchmark data set, using two deep learning architectures, convolutional neural networks and Long Short-Term Memory networks, showing which and when data augmentation methods bring accuracy improvements.

Mobile keystroke dynamics for biometric recognition: An overview

The state of the art of keystroke‐dynamics‐based automatic recognition system, exclusively when dealing with mobile devices, for both verification and identification purposes is discussed.

Siamese Networks for Static Keystroke Dynamics Authentication

  • R. GiotA. Rocha
  • Computer Science
    2019 IEEE International Workshop on Information Forensics and Security (WIFS)
  • 2019
The feasibility of using siamese networks to rely on biometric samples provided by other users instead of requesting a new user to provide a large number of enrollment samples is analyzed.

Keystroke Biometrics for Freely Typed Text Based on CNN model

This paper proposes a multi-user keystroke authentication scheme based on CNN model, which can automatically figure out the appropriate features for the model, adjust and optimize the model constantly to further enhance the performance of model.

A review on smartphone usage data for user identification and user profiling

This paper has a broader lens by reviewing the general applications of smartphone data before focusing on smartphone usage data and discusses some possible data sources that can be used in this research topic.

I Know What You Type on Your Phone: Keystroke Inference on Android Device Using Deep Learning

A deep neural network with four hidden layers is proposed as the baseline for this work, which achieves an accuracy of 47% using categorical cross entropy as the accuracy metric.

Evaluation of Deep Learning Models for Person Authentication Based on Touch Gesture

This work investigates the ability of Deep Learning to automatically discover useful features of touch gesture and use them to authenticate the user in dynamic touch authentication system.

Anomaly detection through keystroke and tap dynamics implemented via machine learning algorithms

The implementation of keystroke biometrics to enhance security using machine-learning algorithms on both Windows and Android and implements machine learning on tap dynamics in Android, allowing both training and prediction and overcoming its computational restrictions.



Continuous user identification via touch and movement behavioral biometrics

This work presents SilentSense, a framework to authenticate users silently and transparently by exploiting dynamics mined from the user touch behavior biometrics and the micro-movement of the device caused by user's screen-touch actions.

Dynamic User Verification Using Touch Keystroke Based on Medians Vector Proximity

The proposed user verification system on mobile phones based on behavioral biometric traits which is a keystroke dynamics derived from a touchable keyboard shows promising results in dynamic authentication area.

Continuous mobile authentication using touchscreen gestures

The introduction of FAST (Fingergestures Authentication System using Touchscreen), a novel touchscreen based authentication approach on mobile devices that provides excellent post-login access security, without disturbing the honest mobile users.

Tapprints: your finger taps have fingerprints

The location of screen taps on modern smartphones and tablets can be identified from accelerometer and gyroscope readings, and TapPrints, a framework for inferring the location of taps on mobile device touch-screens using motion sensor data combined with machine learning analysis is presented.

Associative Patterns of Web Browsing Behavior

It is claimed that spatio-temporal patterns of category of website visited by time of access can uniquely characterize and identify users.

Continuous mobile authentication using a novel Graphic Touch Gesture Feature

  • Xi ZhaoTao FengWeidong Shi
  • Computer Science
    2013 IEEE Sixth International Conference on Biometrics: Theory, Applications and Systems (BTAS)
  • 2013
A novel Graphic Touch Gesture Feature (GTGF) to extract the identity traits from the touch traces' movement and pressure dynamics are represented by intensity values and shapes of the GTGF.

Computation of Cryptographic Keys from Face Biometrics

  • A. GohD. Ngo
  • Computer Science, Mathematics
    Communications and Multimedia Security
  • 2003
We outline cryptographic key-computation from biometric data based on error-tolerant transformation of continuous-valued face eigenprojections to zero-error bitstrings suitable for cryptographic

You're driving and texting: detecting drivers using personal smart phones by leveraging inertial sensors

TEXIVE is able to successfully detect dangerous operations with good sensitivity, specificity and accuracy and achieves the goal of distinguishing drivers and passengers, and detecting texting operations during driving utilizing irregularities and rich micro-movements of users.

Iris recognition for biometric identification using dyadic wavelet transform zero-crossing

A novel biometric identification approach based on the human iris pattern is proposed, to represent the features of the iris by fine-to-coarse approximations at different resolution levelsbased on the discrete dyadic wavelet transform zero-crossing representation.

SigPID: significant permission identification for android malware detection

SIGPID, a malware detection system based on permission analysis to cope with the rapid increase in the number of Android malware, applies 3-level pruning by mining the permission data to identify only significant permissions that can be effective in distinguishing benign and malicious apps.