Abstract
This study introduces two novel hybrid nature-inspired optimization algorithms designed to enhance artificial neural network (ANN) performance in crop recommendation, leveraging remote sensing data from Landsat 8 and 9 platforms. The first hybrid approach combines the Gravitational Search Algorithm (GSA) with the Hunger Games Search (HGS) algorithm, promoting an improved balance between exploration and exploitation through gravitational dynamics and competitive resource-seeking strategies. The second hybrid integrates Electric Eel Foraging Optimization (EEFO) with Crested Porcupine Optimization (CPO), leveraging the foraging adaptability of electric eels and the defensive spatial strategies of crested porcupines to refine search efficiency and convergence stability. These hybrid algorithms were applied to classify crops across Kharif and Rabi seasons in Rajasthan, India. Experimental results reveal that the GSA-HGS hybrid achieves classification accuracies of 95.32% for Kharif and 94.99% for Rabi seasons, while the EEFO-CPO hybrid attains 94.09% and 94.94%, respectively. These findings demonstrate the potential of bio-inspired optimization strategies to support intelligent crop recommendation systems and advance precision agriculture practices in data-scarce agricultural regions.