Abstract :

The Efficiency And Reliability Of Know Your Customer (KYC) Verification Remain Critical Challenges In Modern Banking Due To Redundant Data Collection, Privacy Concerns, And Inconsistent Credit Evaluation. The Existing Blockchain-based KYC Systems Provide Transparency And Immutability But Still Face Limitations In Scalability, Interoperability, And Intelligent Risk Assessment. To Address These Issues, This Paper Proposes An AI-integrated Federated Blockchain KYC Framework That Combines Federated Learning, Zero-knowledge Proofs (ZKP), And Hybrid Blockchain Architecture To Enable Secure, Privacy-preserving, And Intelligent Credit Allocation. In The Proposed Approach, Individual Banks Locally Train Machine Learning Models On Customer Data And Share Only Model Parameters Through The Blockchain, Thereby Maintaining Data Confidentiality. A Hybrid Network Leveraging Hyperledger Fabric And Ethereum Ensures Efficient Transaction Validation And Regulatory Auditability. Smart Contracts Are Designed To Manage User Consent, Automate Verification, And Facilitate Real-time Data Sharing Across Institutions. The System Enhances Scalability, Reduces Redundant KYC Processes, And Supports Dynamic Credit Scoring Through Decentralized Intelligence. Experimental Analysis Demonstrates Improved Data Privacy, Faster Credit Validation, And Reduced Operational Overhead Compared To Conventional Blockchain-only Models. The Proposed Solution Contributes A Secure, Interoperable, And Intelligent KYC Infrastructure Aligned With Evolving Regulatory And Technological Landscapes.The Rapid Expansion Of Cloud Computing Has Significantly Increased The Exposure Of Virtualized Infrastructures To Sophisticated Cyber Threats, Necessitating Efficient And Lightweight Intrusion Detection Mechanisms. This Study Proposes An Optimized Machine Learning–based Intrusion Detection Framework Specifically Designed For Cloud Environments. Unlike Conventional Systems That Rely On High-dimensional Feature Spaces And Computationally Intensive Deep Learning Models, The Proposed Approach Integrates Strategic Feature Engineering With A Random Forest (RF) Classifier To Enhance Detection Efficiency While Minimizing Processing Overhead. The Framework Performs Structured Data Preprocessing, Categorical-tonumerical Transformation, And Inconsistency Elimination Prior To Feature Reduction. A Visualization-driven Feature Selection Strategy Is Employed To Isolate The Most Discriminative Attributes, Enabling The Classifier To Operate On A Compact Yet Highly Informative Feature Subset. This Dimensionality Reduction Not Only Decreases Execution Time But Also Mitigates Overfitting And Improves Generalization Across Datasets. The RF Model Is Trained To Distinguish Between Normal And Anomalous Cloud Traffic, Delivering Strong Performance In Terms Of Accuracy, Precision, And Detection Reliability. By Balancing Computational Efficiency And Detection Capability, The Proposed System Offers A Scalable And Cost-effective Solution For Real-time Cloud Intrusion Detection, Making It Suitable For Dynamic And Resource-constrained Cloud Infrastructures.

Published:

15-2-2026

Issue:

Vol. 26 No. 2 (2026)

Page Nos:

292-298

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite