Continuous adaptive trust framework for enhancing authentication using real-time user behaviour analytics

Abstract

Traditional authentication systems struggle to address the dynamic nature of modern cyber threats, often relying on static rules or historical data that fail to adapt to real-time risks. This research proposes a Framework for Continuous Adaptive Trust (CAT) designed to enhance adaptive authentication by integrating real-time user behavior analytics. The framework dynamically assesses contextual factors—including login time, geolocation, device type, and access patterns—to construct behavioral baselines, detect anomalies through hybrid statistical and machine learning models, and enforce adaptive authentication policies. By leveraging a weighted trust score that combines behavioral analytics with 𝑇𝑡𝑜𝑡𝑎𝑙multi-factor authentication (MFA) outcomes, the system aims to balance security and usability. Integration with the WSO2 Identity Server demonstrates feasibility for enterprise Identity and Access Management (IAM) systems. This work addresses critical gaps in adaptive authentication by prioritizing real-time adaptability, scalability, and privacy-conscious design, offering a foundation for resilient cybersecurity solutions in evolving threat landscapes.