Hybrid AI Models for Real-Time lot Data Analytics on Edge-Cloud Platforms
Keywords:
hybrid AI, edge computing, cloud computing, IoT, realtime analytics, federated learningAbstract
The rapid growth of the Internet of Things (IoT) has led to massive volumes of distributed, high-velocity data that require low-latency, privacy-preserving, and scalable analytics. Pure cloud-based processing often fails to meet real-time constraints, while pure edge analytics struggle with limited computation and storage resources. Hybrid AI, which combines edge and cloud intelligence, has emerged as a promising paradigm for real-time IoT data analytics. This paper proposes a conceptual framework for hybrid AI models on edge–cloud platforms, where latency-sensitive inference is executed at the edge while model training, global optimization, and crossdomain analytics are offloaded to the cloud. We review recent advances in edge and hybrid architectures, discuss AI deployment patterns, and present a reference architecture with a layered data flow. A comparative table highlights key trade-offs between cloud-only, edge-only, and hybrid AI strategies. The paper concludes with open challenges in orchestration, resource management, security, and standardization.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Author(s)

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