Enhancing Online Exam Outcome Dependability Through a Blockchain-Based Framework for Secure and Transparent Assessment

Abstract

Online examinations that companies rely on more frequently have made traditional centralized Learning Management Systems (LMS) vulnerable to security threats and authentication problems and result manipulation issues. The place of storing examination data in a central location creates risks for intentional changes which compromises the process fairness as well as integrity. Currently deployed blockchain solutions are ineffective because they fail to deliver both economical solutions and scalable systems that work well with AI proctoring functions. This study introduces the BlockchainBased Examination Framework (BEF) as an integrated system which unites multi-LMS operation with blockchain-based safe storage along with AI-powered examination surveillance features for real-time academic dishonesty discovery. The system utilizes Ethereum together with Hyperledger Fabric and Solana blockchains to guarantee result security and activates Zero-Knowledge Proofs (ZKP) and ECDSA signatures for authentication privacy and implements AI models for live examination monitoring. A thorough examination analyzed speed and scalability and financial efficiency together to evaluate these aspects of the three platform frameworks. Solana demonstrates superior performance through its$\mathbf{6 5, 0 0 0}$Transactions Per Second along with its affordable transaction fee of $0.00025 that makes it the best scalable and efficient choice. The AI-proctoring system demonstrated a 97.8% accuracy level together with a$\mathbf{2. 2 \%}$false positive error rate which improved examination security. The research demonstrates blockchain implementation as a critical enhancement for exam security as well as transparency levels. The future project will concentrate on Ethereum Layer-2 scaling alongside deep learning improvements to AI proctoring systems for better cost reduction and flexibility.