<![CDATA[The rapid growth of the e-commerce ecosystem has introduced complex cybersecurity threats, particularly account takeovers and transaction fraud. Traditional static authentication and fragmented security modules are no longer sufficient to mitigate these dynamic risks. This research aims to design, implement, and evaluate an integrated security architecture that combines adaptive Risk-Based Authentication with a real-time Order Risk Engine. Utilizing an experimental approach within the Secure Software Development Life Cycle framework, a server-side rendered prototype was developed and subjected to synthetic anomaly injections in an isolated local testbed. The system evaluates operational contexts, such as unfamiliar IP addresses, bruteforce attempts, and abnormal order velocities, using a deterministic scoring mechanism to trigger automated interventions ranging from multi-factor authentication challenges to absolute access blocks. The empirical findings demonstrate that the proposed end-to-end risk scoring engine achieved a zero percent false-positive rate for legitimate users while successfully mitigating all simulated critical threats and account takeover attempts. Furthermore, the integration of stateless session management maintained an exceptionally low computational latency, ensuring a seamless user experience. These results imply that a unified risk-scoring model provides a highly effective, autonomous, and scalable blueprint for securing modern e-commerce platforms against multi-layered exploitations.]]>
Copyrights © 2026
