<p>Increasingly, fog computing is being used by Internet of Things (IoT) applications to do computations at the network edge of devices instead of cloud computing. The low latency of fog computing, as opposed to the high latency of the cloud, delay-sensitive applications can benefit from improved quality of service. To increase service quality and reaction time in IoT applications, workloads dispersed efficiently between fog nodes. Therefore, this research proposes a novel model of Enhanced Task Distribution and Adaptive Resource Management utilizing Spike-Induced Graph Neural Networks with Optimized Offloading Mechanisms in Fog Computing for Improved Efficiency (TDARM-SIGNN-FC). To predict the optimal fog node for task offloading, task attributes and fog node characteristics are analyzed to ensure efficient resource management. Here, fog node prediction is performed using Spike-Induced Graph Neural Networks (SIGNN) to determine which node is best suited for offloading. The predicted node is then fed into the Starfish Optimization Algorithm (SFOA), which efficiently manages the task offloading process. Then the proposed TDARM-SIGNN-FC is implemented and the performance metrics like Energy Consumption (EC), Total Network Usage (TNU), Execution Time (ET), Average Response Time (ART), Standard Deviation (SD) and Cost are analyzed. Finally the performance of proposed TDARM-SIGNN-FC technique provides 24.65%, 21.95% and 29.56% lower energy consumption, 32.05%, 23.99% and 19.56% lower total network usage and 24.78%, 28.97% and 29.45% lower execution time while compared with existing methods like Prioritized Task offloading mechanism in Cloud-Fog Computing using improved Asynchronous Advantage Actor Critic Algorithm (PTOM-CFC-ACA), Hybrid Task scheduling technique in fog computing using fuzzy logic and Deep Reinforcement learning (TST-FC-DLR), and energy-efficient task offloading in fog computing for 5G cellular network (ETO-FC-5GCN) respectively.</p>