Unified Blockchain and Machine Learning Framework for Secure Digital Certification and Adaptive Course Recommendations

Abstract

This growth of digital learning platforms has presented a twin need: to deliver a learner a highly personalized educational journey and to deliver academic credentials that are verifiable, safe, and unchangeable. The current systems tend to address these goals separately and as a result, there are disjointed ecosystems with complex recommendation engines without trusted credentialing systems and sound certification systems that do not provide any course selection guidance. To fill this gap, this paper presents the Integrated Adaptive Learning and Certification Framework (IALCF), a new architecture that integrates into a LightGBM-based recommendation system a blockchain-based digital certification protocol in a synergistic manner. The recommendation module is an active learner profile analyzer that uses past performance, real-time interaction metrics and dynamically recommenders, predicting course selection with an accuracy of 98.7 and mean absolute error (MAE) of 1.18. The certification module is based on a more advanced X.509 standard with a delegated Proof-of-Stake (dPoS) blockchain, which forms a tamper-evident credential storage and an efficient verification algorithm, which has a verification success rate of over 95 percent in high-load conditions. The experimental findings reveal that the IALCF is a scalable, efficient and safe end-to-end solution to contemporary e-learning settings and is effective in integrating personalized learning with credible management of credentials.