Decentralized Edge Intelligence for Big Data Analytics-Assisted E-Learning

Edge Intelligence

The convergence of AI and Edge Computing has given rise to Edge Intelligence, which uses the computer and communication capabilities of end devices and edge servers to process data closer to where it is produced, allowing for large-scale and efficient AI deployment. The privacy-preserving machine learning paradigm known as Federated Learning (FL), which allows data owners to conduct model training without submitting their raw data to third-party servers, is one of the enabling technologies of Edge Intelligence. On the other hand, the FL network is expected to include thousands of heterogeneous dispersed devices. As a result, ineffective communication remains a significant obstacle (Dahdouh et al., 2020). The Hierarchical Federated Learning (HFL) framework has been developed to mitigate node failures and device dropouts by designating cluster chiefs to assist data owners through intermediate model aggregation. The reliance on a central controller, such as the model owner, is reduced with this decentralized learning strategy.

Intelligent personal assistants, context-related and personalized purchase advice, automated video monitoring, and smart appliances are just a few of the intelligent applications that have emerged from recent breakthroughs in AI. Such applications swiftly rose to prominence and garnered significant popularity in a short period because they improved the productivity and overall efficiency of human-based activities, positively impacting people's lives. With the rise of mobile, pervasive computing, and the Internet of Things, the cloud's position as the greatest data concentrator is eroding. Given the phenomenon's scale, transferring massive amounts of data to a wholly centralized AI is inconceivable. The edge ecosystem is critical for developing intelligent apps and systems that begin at the edge and interface with centralized AI systems.

E-learning and intelligent education are developing fields that allow the rapid integration of smart technologies, smart environments, and smart learning and teaching methods. Teachers and students benefit from the advantages of cloud computing, the primary paradigm on which the future of education is based, to deliver more effective learning content within an integrated environment.