TypeNet: Deep Learning Keystroke Biometrics

• URL: http://arxiv.org/abs/2101.05570v2
• Date: Thu, 18 Feb 2021 17:40:57 GMT
• Title: TypeNet: Deep Learning Keystroke Biometrics
• Authors: Alejandro Acien and Aythami Morales and John V. Monaco and Ruben Vera-Rodriguez and Julian Fierrez
• Abstract summary: We introduce TypeNet, a Recurrent Neural Network trained with a moderate number of keystrokes per identity. With 5 gallery sequences and test sequences of length 50, TypeNet achieves state-of-the-art keystroke biometric authentication performance. Our experiments demonstrate a moderate increase in error with up to 100,000 subjects, demonstrating the potential of TypeNet to operate at an Internet scale.
• Score: 77.80092630558305
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: We study the performance of Long Short-Term Memory networks for keystroke biometric authentication at large scale in free-text scenarios. For this we introduce TypeNet, a Recurrent Neural Network (RNN) trained with a moderate number of keystrokes per identity. We evaluate different learning approaches depending on the loss function (softmax, contrastive, and triplet loss), number of gallery samples, length of the keystroke sequences, and device type (physical vs touchscreen keyboard). With 5 gallery sequences and test sequences of length 50, TypeNet achieves state-of-the-art keystroke biometric authentication performance with an Equal Error Rate of 2.2% and 9.2% for physical and touchscreen keyboards, respectively, significantly outperforming previous approaches. Our experiments demonstrate a moderate increase in error with up to 100,000 subjects, demonstrating the potential of TypeNet to operate at an Internet scale. We utilize two Aalto University keystroke databases, one captured on physical keyboards and the second on mobile devices (touchscreen keyboards). To the best of our knowledge, both databases are the largest existing free-text keystroke databases available for research with more than 136 million keystrokes from 168,000 subjects in physical keyboards, and 60,000 subjects with more than 63 million keystrokes acquired on mobile touchscreens.

Related papers

• TapType: Ten-finger text entry on everyday surfaces via Bayesian inference [32.33746932895968]
  TapType is a mobile text entry system for full-size typing on passive surfaces. From the inertial sensors inside a band on either wrist, TapType decodes and relates surface taps to a traditional QWERTY keyboard layout.
  arXiv  Detail & Related papers  (2024-10-08T12:58:31Z)
• Blind-Touch: Homomorphic Encryption-Based Distributed Neural Network Inference for Privacy-Preserving Fingerprint Authentication [8.61984933438984]
  Homomorphic encryption allows computations encrypted on data without decrypting. Blind-Touch can keep fingerprint data encrypted on the server while performing machine learning operations. Blind-Touch achieves high accuracy on two benchmark fingerprint datasets.
  arXiv  Detail & Related papers  (2023-12-18T09:05:34Z)
• OverHear: Headphone based Multi-sensor Keystroke Inference [1.9915929143641455]
  We develop a keystroke inference framework that leverages both acoustic and accelerometer data from headphones. We achieve top-5 key prediction accuracy of around 80% for mechanical keyboards and around 60% for membrane keyboards. Results highlight the effectiveness and limitations of our approach in the context of real-world scenarios.
  arXiv  Detail & Related papers  (2023-11-04T00:48:20Z)
• TypeFormer: Transformers for Mobile Keystroke Biometrics [11.562974686156196]
  We propose a novel Transformer architecture to model free-text keystroke dynamics performed on mobile devices for the purpose of user authentication. TypeFormer outperforms current state-of-the-art systems achieving Equal Error Rate (EER) values of 3.25% using only 5 enrolment sessions of 50 keystrokes each.
  arXiv  Detail & Related papers  (2022-12-26T10:25:06Z)
• Mobile Behavioral Biometrics for Passive Authentication [65.94403066225384]
  This work carries out a comparative analysis of unimodal and multimodal behavioral biometric traits. Experiments are performed over HuMIdb, one of the largest and most comprehensive freely available mobile user interaction databases. In our experiments, the most discriminative background sensor is the magnetometer, whereas among touch tasks the best results are achieved with keystroke.
  arXiv  Detail & Related papers  (2022-03-14T17:05:59Z)
• TapNet: The Design, Training, Implementation, and Applications of a Multi-Task Learning CNN for Off-Screen Mobile Input [75.05709030478073]
  We present the design, training, implementation and applications of TapNet, a multi-task network that detects tapping on the smartphone. TapNet can jointly learn from data across devices and simultaneously recognize multiple tap properties, including tap direction and tap location.
  arXiv  Detail & Related papers  (2021-02-18T00:45:41Z)
• On the Inference of Soft Biometrics from Typing Patterns Collected in a Multi-device Environment [47.37893297206786]
  In this paper, we study the inference of gender, major/minor, typing style, age, and height from the typing patterns collected from 117 individuals in a multi-device environment. For classification tasks, we benchmark the performance of six classical machine learning (ML) and four deep learning (DL) classifiers. The results are promising considering the variety of application scenarios that we have listed in this work.
  arXiv  Detail & Related papers  (2020-06-16T20:25:58Z)
• Keystroke Biometrics in Response to Fake News Propagation in a Global Pandemic [77.79066811371978]
  This work proposes and analyzes the use of keystroke biometrics for content de-anonymization. Fake news have become a powerful tool to manipulate public opinion, especially during major events.
  arXiv  Detail & Related papers  (2020-05-15T17:56:11Z)
• TypeNet: Scaling up Keystroke Biometrics [79.19779718346128]
  We first analyze to what extent our method based on a Recurrent Neural Network (RNN) is able to authenticate users when the amount of data per user is scarce. With 1K users for testing the network, a population size comparable to previous works, TypeNet obtains an equal error rate of 4.8%. Using the same amount of data per user, as the number of test users is scaled up to 100K, the performance in comparison to 1K decays relatively by less than 5%.
  arXiv  Detail & Related papers  (2020-04-07T18:05:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.