Edge-AI in IoT: Leveraging Cloud Computing and Big Data for Intelligent Decision-Making

Abstract

The extremely rapid increase in the number of these IoT devices has led to an unprecedented creation of data that requires intelligent and efficient mechanisms for decision-making. Today, Edge Artificial Intelligence (Edge-AI) is transforming the world with real-time data-processing capabilities, minimizing latency, optimizing bandwidth, and establishing separation for security. The integration of Edge-AI, cloud computing, and big data technology is studied in this research to optimize intelligent decision-making in IoT ecosystems. Using the distributed nature of edge computing, we present an Edge-Cloud AI framework which dynamically assigns computation workloads onto the edge nodes and centralized cloud infrastructures. The experimental results together with the validation through real IoT scenarios show that the proposed methods outstand in terms of response time, energy consumption and predictiveness. This approach strikes a proper balance between inference at the edge in real time and training of the models and big data analytics at the cloud; thus allowing adoption in intelligent solutions leading to adaptive, context aware intelligence. Innovative Elements include a new decision-making model based on federated learning, distributed pre-processing of data, and mechanisms for maintaining the confidentiality of data. These results highlight the capabilities of Edge-AI in improving the scalability and reliability of IoT applications in different domains such as smart cities, healthcare, and industrial automation. This work lays the groundwork for the next step in autonomous IoT, connecting edge intelligence with cloud-based learning analytics.