In this study, we propose an enhanced and scalable implementation of the K-Means clustering algorithm tailored for large-scale datasets. By integrating the Mini-Batch K-Means with distributed computing frameworks such as Apache Spark, we improve clustering performance in terms of speed and convergence. Additionally, we conduct a comparative analysis with density-based spatial clustering of applications with noise, GMM, and Mean Shift to evaluate the trade-offs in terms of accuracy and computational cost. Our experiments on real-world customer segmentation data demonstrate that the proposed approach achieves high clustering efficiency while maintaining scalability. The optimal number of clusters is determined using the Elbow method and Silhouette coefficient, with visualization confirming improved segmentation quality. This framework supports more effective decision-making in big data environments.