The importance of computing and storage capacity has increased over time, and the importance of data mining in industrial engineering has become more apparent. Recently, artificial intelligence and machine learning have made significant advancements in industrial engineering. Federated learning is a machine learning technique that aims to solve the problem of distributed computing systems and their applications of data storage while ensuring data privacy. Tolpegin et al. conducted research on data poisoning attacks in a federated learning system, which we have extended to an analysis of the efficiency of Tolpegin's proposed defense technique. We have subsequently compared the efficiency using uniform manifold approximation and projection (UMAP), principal component analysis (PCA), kernel PCA (KPCA), and K-means clustering algorithms. This study confirms that UMAP performs better than PCA, KPCA, and K-means, and provides excellent performance in mitigating data-poisoning attacks.