Title: Keystroke dynamics and quantum machine learning

Authors: Namisha Bhasin; Sanjay Kumar Sharma; Rajesh Mishra

Addresses: Gautam Buddha University, Greater Noida, 201312, Uttar Pradesh, India ' Gautam Buddha University, Greater Noida, 201312, Uttar Pradesh, India ' Gautam Buddha University, Greater Noida, 201312, Uttar Pradesh, India

Abstract: The performance of machine learning algorithms is often suboptimal in identifying and classifying patterns hence, there was always a requirement for methods that could provide optimal solutions. Quantum algorithms have demonstrated a significantly greater efficiency in many tasks than traditional machine learning algorithms. Quantum computers leverage unique properties such as entanglement and superposition, allowing them to generate patterns inaccessible to classical systems. Keystroke dynamics, a method for user identification based on typing style, is categorised into static authentication, where users input a username/password combination of 15-20 letters, and dynamic authentication, where users type unbiased text such as emails, chats, or online exams. Both static and dynamic authentication primarily involve involuntary actions. This research paper focuses on authenticating users based on static keystroke dynamics using various quantum and hybrid algorithms.

Keywords: quantum support vector classifier; QSVC; ZZFeatureMap; quantum neural network; QNN; variational quantum circuit; VQC.

DOI: 10.1504/IJBM.2025.143727

International Journal of Biometrics, 2025 Vol.17 No.1/2, pp.132 - 150

Received: 05 Dec 2023
Accepted: 15 Mar 2024
Published online: 06 Jan 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article