<p>Swarm intelligence algorithms are key tools for solving complex optimization problems. The Sparrow Search Algorithm (SSA) is a novel swarm intelligence algorithm, but it has problems such as insufficient convergence speed, low solution accuracy, and poor stability. To overcome these disadvantages, we propose an enhanced SSA by incorporating the Whale Optimization Algorithm (WOA), which innovatively integrates three aspects improvement strategies: (1) an initialization strategy based on chaotic mapping and opposition-based learning is used to improve the diversity of the initial population, (2) a spiral updating strategy based on the novel inverted S-shaped inertia weight is used to balance the global exploration and local exploitation capabilities of ESSA, thereby improving the convergence speed, solution accuracy, and stability of ESSA, and (3) a position mutation strategy based on double random numbers is used to perturb the global optimal solution, thereby avoiding ESSA from falling into premature convergence. The superiority and competitiveness of ESSA were verified through comprehensive numerical experiments with 12 cutting-edge optimization algorithms on CEC2017 and CEC2022. In addition, ablation experiments have demonstrated the independent contribution of various improvement strategies of ESSA. We also use ESSA to optimize support vector machines (SVM) and propose ESSA-SVM to solve feature selection problems, where ESSA is used for feature selection and the SVM optimized by ESSA is used for classification of the dataset. The experimental results on the datasets downloaded from UCI and Kaggle platforms show that ESSA-SVM has high superiority and stability compared to the other 12 efficient feature selection algorithms. These results indicate that ESSA has great potential in a wide range of practical optimization scenarios.</p>