This research introduces the S-shaped Grey Wolf Optimizer-based FOX (FOX-GWO) algorithm for feature selection, which enhances the original FOX algorithm by integrating Grey Wolf Optimization (GWO) and an S-shaped transfer function. The FOX-GWO algorithm improves the local search capability of the FOX algorithm and enables efficient conversion of continuous values to binary values, making it suitable for feature selection tasks. The algorithm was tested on 18 datasets and outperformed other optimization algorithms in terms of average accuracy (83.33%), reduced feature dimensionality (61.11%), and average fitness value (72.22%). The results indicate that FOX-GWO is effective in exploring high-dimensional spaces and has practical value in complex data analysis. The algorithm's performance was evaluated using various metrics, including fitness values, accuracy, and computational efficiency. The study also highlights the importance of balancing exploration and exploitation in optimization problems and suggests future research directions, such as developing a parallel version of the algorithm to reduce computational time. The proposed FOX-GWO algorithm demonstrates superior performance in feature selection tasks and offers a robust solution for high-dimensional data analysis.This research introduces the S-shaped Grey Wolf Optimizer-based FOX (FOX-GWO) algorithm for feature selection, which enhances the original FOX algorithm by integrating Grey Wolf Optimization (GWO) and an S-shaped transfer function. The FOX-GWO algorithm improves the local search capability of the FOX algorithm and enables efficient conversion of continuous values to binary values, making it suitable for feature selection tasks. The algorithm was tested on 18 datasets and outperformed other optimization algorithms in terms of average accuracy (83.33%), reduced feature dimensionality (61.11%), and average fitness value (72.22%). The results indicate that FOX-GWO is effective in exploring high-dimensional spaces and has practical value in complex data analysis. The algorithm's performance was evaluated using various metrics, including fitness values, accuracy, and computational efficiency. The study also highlights the importance of balancing exploration and exploitation in optimization problems and suggests future research directions, such as developing a parallel version of the algorithm to reduce computational time. The proposed FOX-GWO algorithm demonstrates superior performance in feature selection tasks and offers a robust solution for high-dimensional data analysis.