Swarm Optimization-Based Hyperparameter Selection for Machine Learning Algorithms in Indoor Localization

In this study, we compare multiple machine learning algorithms for indoor positioning applications, offering insights into the application of swarm optimization algorithms for hyperparameter selection in indoor positioning tasks. The study's findings demonstrate that both Particle S warm Optimization (PSO) and Whale Optimization Algorithms (WOA) improve the performance of machine learning models. Specifically, Random Forest (RF)-based classification demonstrates the highest accuracy, precision, and recall. The results also indicate that an increased number of access points results in improved performance and decreased standard deviations. The study contributes to the optimization and refinement of machine learning algorithms for indoor localization, opening avenues for more accurate and reliable positioning systems.