BEHAVIORAL BIOMETRIC CONTINUOUS USER AUTHENTICATION

Fuente: Zenodo
Salvato in:
Dettagli Bibliografici
Autore principale: Mrs. Bhavana, Jagati Goutham, Jakkula Karnakar, Javidi Bharath Reddy, and Nommula Anuroop
Natura: Recurso digital
Pubblicazione: Zenodo 2026
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866901775291777024
author Mrs. Bhavana, Jagati Goutham, Jakkula Karnakar, Javidi Bharath Reddy, and Nommula Anuroop
author_facet Mrs. Bhavana, Jagati Goutham, Jakkula Karnakar, Javidi Bharath Reddy, and Nommula Anuroop
contents <p><a href="https://ijetrm.com/issues/files/Apr-2026-25-1777092517-BEHAVIORAL-APR67-2026.pdf"><em><strong>Modern digital systems</strong></em></a> rely heavily on secure access control, especially in environments where sensitive<br>information is processed. Conventional login-based methods are no longer sufficient because they validate<br>identity only once and can be bypassed through credential compromise or session takeover attacks. This<br>limitation creates a need for mechanisms that can verify user identity continuously during system usage.<br>The approach presented in this work focuses on evaluating how a user interacts with the system instead of<br>relying only on static credentials. Interaction patterns such as typing rhythm and cursor movement are<br>monitored over time and treated as behavioral signatures. These patterns are analyzed as sequential data,<br>allowing the system to recognize consistency or deviation in user activity.<br>A sequence-based learning model is used to interpret these interaction streams and generate a confidence<br>measure representing user authenticity. The system operates passively in the background and evaluates data at<br>regular intervals. When a mismatch is detected, an additional verification step is triggered to confirm identity. If<br>the verification fails, access is restricted by ending the active session and limiting application usage. This<br>layered mechanism improves security by combining continuous observation with adaptive decision-making.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_19753242
institution Zenodo
language
publishDate 2026
publisher Zenodo
record_format zenodo
spellingShingle BEHAVIORAL BIOMETRIC CONTINUOUS USER AUTHENTICATION
Mrs. Bhavana, Jagati Goutham, Jakkula Karnakar, Javidi Bharath Reddy, and Nommula Anuroop
<p><a href="https://ijetrm.com/issues/files/Apr-2026-25-1777092517-BEHAVIORAL-APR67-2026.pdf"><em><strong>Modern digital systems</strong></em></a> rely heavily on secure access control, especially in environments where sensitive<br>information is processed. Conventional login-based methods are no longer sufficient because they validate<br>identity only once and can be bypassed through credential compromise or session takeover attacks. This<br>limitation creates a need for mechanisms that can verify user identity continuously during system usage.<br>The approach presented in this work focuses on evaluating how a user interacts with the system instead of<br>relying only on static credentials. Interaction patterns such as typing rhythm and cursor movement are<br>monitored over time and treated as behavioral signatures. These patterns are analyzed as sequential data,<br>allowing the system to recognize consistency or deviation in user activity.<br>A sequence-based learning model is used to interpret these interaction streams and generate a confidence<br>measure representing user authenticity. The system operates passively in the background and evaluates data at<br>regular intervals. When a mismatch is detected, an additional verification step is triggered to confirm identity. If<br>the verification fails, access is restricted by ending the active session and limiting application usage. This<br>layered mechanism improves security by combining continuous observation with adaptive decision-making.</p>
title BEHAVIORAL BIOMETRIC CONTINUOUS USER AUTHENTICATION
url https://doi.org/10.5281/zenodo.19753242