Disputation: Kamer Vishi

Doctoral candidate Kamer Vishi at the Department of Informatics, Faculty of Mathematics and Natural Sciences, is defending the thesis Security and Privacy in User Authentication: Aspects of Fusion, Machine Learning, and Privacy in Biometric Authentication for the degree of Philosophiae Doctor.

    Picture of the candidate

    Photo: UiO

     

    The PhD defence will be partially digital, in Kristen Nygaards sal (5370), Ole-Johan Dahls hus and streamed directly using Zoom. The host of the session will moderate the technicalities while the chair of the defence will moderate the disputation.

    Ex auditorio questions: the chair of the defence will invite the attending audience at Kristen Nygaards sal to ask ex auditorio questions. 

    Trial lecture

    Performance Comparison of Algorithms for Soft Biometric Keystroke Dynamics for Age and Gender Detection

    Time and place: March 6,  2023 10:15 AM, Kristen Nygaards sal (5370), Ole-Johan Dahls hus/ZOOM

    Main research findings

    The research in this PhD dissertation focuses on investigating user authentication's accuracy, assurance, and security improvements by introducing high-quality samples and the fusion of biometric modalities while considering privacy implications. The research is based on four research questions, which aim to understand the impact of information fusion, the quality of biometric samples, machine learning, and privacy protection aspects on the recognition accuracy of a biometric system. The research is divided into four main areas:

    1. A new approach for multi-biometric fusion based on subjective logic is proposed, and its impact on recognition accuracy is evaluated.

    2. The quality of biometric samples is analyzed, and its impact on recognition accuracy is studied.

    3. The use of machine learning in biometric authentication systems is examined, and its impact on recognition accuracy is assessed.

    4. Privacy protection aspects are examined and integrated into a biometric authentication system to enable compliance with GDPR. The findings of this research have been published in reputed journals and conferences and contribute to the field of biometric authentication by significantly increasing the accuracy and reliability of authentication while also addressing privacy concerns.

    Ultimately, this work provides a comprehensive approach to improving the security and privacy of biometric authentication systems.

    Image may contain: Font, Parallel, Diagram, Brand, Terrestrial plant.

     

    Adjudication committee:

     

    • Professor Patrick Bours, NTNU i Gjøvik , Norway
    • Associate Professor Estelle Cherrier, ENSICAEN, GREYC, France
    • Associate Professor Silvia Masiero, University of Oslo, Department of informatics

    Supervisors

    • Professor Audun Jøsang, Department of Informatics, UIO, Norway
    • Assistant Professor Magdalena Ivanovska, Department of Data Science and Analytics, BI Norwegian Business School

    • Professor Mohammad Derawi, Department of Electronic Systems, Faculty of Information Technology and Electrical Engineering, NTNU i Gjøvik

    Chair of defence:

    Professor Dag Langmyhr

    Candidate contact information

    Contact information at Department: Mozhdeh Sheibani Harat 

    Published Feb. 20, 2023 11:37 AM - Last modified Mar. 2, 2023 2:59 PM