Secure online examination in academic institutions has long depended on dedicated software solutions such as Safe Exam Browser (SEB) to enforce assessment integrity. While effective in controlled environments, such tools introduce substantial deployment barriers including mandatory installation on student devices, operating system incompatibilities, and documented failure rates exceeding 30% in live examination sessions. This paper proposes and evaluates a browser-native examination security framework that replicates the essential security guarantees of kiosk-mode examination software using only standardized web APIs1/specifically the HTML5 Fullscreen API and the W3C Page Visibility API1/without requiring any software installation on student machines. The framework is integrated into a complete web-based examination platform supporting role-based access control, difficulty-stratified randomized question distribution, attendance-verified quiz eligibility, and real-time automated result computation. The system is implemented using PHP, MySQL, HTML5, CSS3, and vanilla JavaScript on a standard LAMP server environment. Evaluation through 53 structured test cases across functional, security, and accuracy dimensions yielded a 100% pass rate. A comparative analysis against Safe Exam Browser demonstrates that the proposed approach achieves equivalent malpractice prevention with significantly lower deployment complexity and zero hardware or operating system constraints.

This paper has presented and evaluated a browser-native framework for secure online examination that eliminates dependency on external examination software while achieving functional equivalence with dedicated security tools for the malpractice prevention scenarios most relevant to institutional academic assessment. The framework leverages the HTML5 Fullscreen API for screen lockdown enforcement and the W3C Page Visibility API for tab-switch detection, combining these with complementary content restrictions and a multi-layered access control architecture to create a defense-in-depth security posture operable within any modern web browser. The integration of difficulty-stratified randomized question distribution addresses the complementary problem of inter-student answer sharing, while the automated result computation engine eliminates the delays and errors associated with manual evaluation. Post-examination administrative mechanisms including answer key correction with automatic result recalculation represent functional contributions not commonly found in existing web-based examination systems. Empirical evaluation through 53 structured test cases demonstrated 100% correctness and security compliance. Comparative analysis against Safe Exam Browser confirms that the proposed framework achieves equivalent security coverage across the principal malpractice vectors while eliminating all client-side software requirements. Future work will address the identified limitation in screen recording prevention through integration of the Media Devices API for webcam-based behavioural monitoring, which can detect screen recording application activity through reflection analysis. Additional planned enhancements include server-side password hashing, support for image-embedded questions, mobile interface optimization, and multi-tenant architecture for cross-institutional deployment.

[1] ETH Zurich, “Safe Exam Browser Technical Documentation,” 2023. [Online]. Available: https://safeexambrowser.org/ [2] G. Sindre and A. Vegendla, “E-exams versus paper exams: A comparative analysis of cheating-related security threats and countermeasures,” Proceedings of the Norwegian Information Security Conference (NISK), 2015, pp. 44–55. [3] World Wide Web Consortium, “Fullscreen API Living Standard,” WHATWG, 2022. [Online]. Available: https://fullscreen.spec.whatwg.org/ [4] World Wide Web Consortium, “Page Visibility (Second Edition),” W3C Recommendation, Oct. 2013. [Online]. Available: https://www.w3.org/TR/page-visibility/ [5] I. Atoum, A. Otoom, and A. A. Ali, “A holistic cyber security implementation framework,” Information Management and Computer Security, vol. 25, no. 3, pp. 285–306, 2017. [6] V. B. Rathod and R. S. Kejkar, “Online examination system using client server architecture,” International Journal of Computer Applications, vol. 975, no. 8887, pp. 14–17, 2014. [7] S. N. Ogujiofor and O. I. Abiodun, “Randomized examination question delivery as a malpractice prevention strategy in computer-based testing,” International Journal of Scientific Research in Computer Science, Engineering and Information Technology, vol. 5, no. 2, pp. 718–725, 2019. [8] S. M. Al-Saleem, F. Al-Abed Al-Haq, and N. Alnazzawi, “A secure e-examination system,” International Journal of Emerging Technologies in Learning, vol. 7, no. 3, pp. 34–39, 2012. [9] S. Chaudhary and S. Tripathi, “Browser-based online proctoring system using HTML5 APIs,” International Journal of Advanced Research in Computer Science, vol. 11, no. 3, pp. 45–51, 2020. [10] R. Kumar and V. Sharma, “Automated result generation in academic institutions: A review,” Journal of Educational Technology Systems, vol. 49, no. 3, pp. 312–329, 2021. [11] S. Ssemugabi and R. De Villiers, “Usability of web-based e-learning in higher education: A case study of two systems used at the University of South Africa,” Proceedings of SAICSIT, pp. 327–336, 2010. [12] OWASP Foundation, “OWASP Top Ten Web Application Security Risks,” 2023. [Online]. Available: https://owasp.org/www-project-top-ten/