Risk-adaptive transaction security framework integrating hybrid anomaly detection, post-quantum cryptography and...

Area of Science:

• Cybersecurity and Cryptography
• Financial Technology (FinTech)
• Intelligent Systems

Background:

• Current digital payment fraud detection relies on isolated systems and traditional methods, insufficient for evolving threats and post-quantum security.
• Existing approaches lack holistic trust evaluation and dynamic adaptation to complex fraud patterns.
• Stringent latency and reliability constraints challenge traditional fraud detection in modern payment systems.

Purpose of the Study:

• To propose an end-to-end, risk-adaptive transaction security framework for digital payments.
• To reconceptualize fraud detection as a dynamic, continuous security process.
• To enhance trust, auditability, and long-term security against evolving and post-quantum threats.

Main Methods:

• Hybrid Anomaly-Guided Adaptive Detection (HAGAD) integrating supervised prediction, unsupervised anomaly detection, and temporal analysis.
• Dynamic risk scoring to govern Adaptive Key Ratcheting (AKR), risk-triggered association quorums, and post-quantum threshold signing.
• Selective use of permissioned blockchain and an Adaptive Assurance Scoring (AAS) mechanism for trustworthiness evaluation.

Main Results:

• The framework achieves high reliability with balanced precision and recall.
• Enhanced security mechanisms are activated adaptively based on transaction risk, avoiding overhead for low-risk transactions.
• Demonstrates robust decision validation and efficient resource utilization.

Conclusions:

• The proposed framework provides an integrated, risk-adaptive pipeline for next-generation electronic payment systems.
• It unifies intelligent detection, advanced cryptography, auditability, and assurance evaluation for resilient security.
• Establishes a future-ready foundation for secure digital transactions against emerging threats.