Warthog Optimization Algorithm: A New Nature-Derived Solution for Optimization Challenges

Data plays an increasingly critical role in today’s digital era, driving advancements in Information Technology (IT) and Computer Science (CS). As the volume, complexity, and dimensionality of data have grown manifold, efficient techniques have become the necessity for processing, analyzing, and extracting meaningful data. Challenges such as managing high-dimensional datasets, optimizing distributed resources, and solving complex, nonlinear problems require new adaptable optimization algorithms (OAs) that balance exploration (searching widely for various possible solutions) with exploitation (focusing on and improving the best solutions). Optimization techniques play a critical role across diverse domains including machine learning (ML), engineering design (ED), and image analysis. Traditional algorithms often struggle with the intricacies of modern, complex search spaces, which have led researchers to turn to metaheuristic approaches. Nature-derived optimization techniques, known for their adaptive strategies, have come up as powerful tools in this context. This research introduces the Warthog Optimization Algorithm (WartOA), a novel bio-inspired method that uniquely integrates burrow-based retreat and adaptive foraging strategies for dynamic exploration-exploitation balance. Following evaluation on 23 Benchmark Functions (BFs), multiple ED problems, and feature selection for the NSL-KDD intrusion detection dataset, WartOA demonstrates strong potential and robustness when compared with established optimization techniques.