The widespread adoption of Internet-of-Things (IoT) applications has led to the development of the Fog computing paradigm, which enables the seamless integration of mobile-edge and cloud resources. Efficiently allocating and provisioning application tasks in these environments is challenging due to limited resource capacities, IoT mobility, network hierarchy, resource heterogeneity, and stochastic behaviours. In this interconnected world, the demand for more computing devices, especially for delay-sensitive applications, is crucial. This paper proposes a Hybrid Fog Resource Provisioning (HFRP) algorithm for optimal resource allocation using MKFCM and the Flower Pollination Algorithm (FPA) to efficiently select and provision available resources. The first phase involves discovering and clustering available resources. In the second phase, the FPA is used to allocate tasks. The simulation results of the HFRP algorithm is compared with the Whale Optimization Algorithm-Based Resource Allocation (WORA). Across 15 fog nodes and executing between 100 and 700 tasks, the HFRP algorithm shows a 25.83% average cost reduction compared to WORA.