This paper investigates the performance of various clustering algorithms, DBSCAN, K-Means, and Agglomerative Clustering in high-dimensional data, enhanced through Particle Swarm Optimization (PSO) and machine learning techniques. Through evaluation using multiple clustering metrics, including Silhouette score, Davies-Bouldin score, and Calinski-Harabasz score, DBSCAN demonstrated superior performance in Silhouette and Davies-Bouldin scores, while Agglomerative Clustering excelled in the Calinski-Harabasz score. The study further explores linkage methods, revealing that the Ward linkage in Agglomerative Clustering performs optimally when the number of data points is larger. Additional evaluation with metrics like AMI and NMI confirmed DBSCAN’s overall performance. The findings highlight the importance of choosing the right algorithm and linkage method for clustering high-dimensional data effectively.