Data mining has become an essential tool in extracting valuable insights from vast repositories of data across various domains. The objective of this study is to develop enhanced clustering techniques that are optimized for large-scale data mining applications. We propose a hybrid algorithm that integrates density-based and partition-based methods to improve computational efficiency and accuracy. The algorithm was tested on multiple datasets obtained from real-world applications, demonstrating significant improvements in cluster purity and computational time. Our findings suggest that the hybrid approach not only maintains the integrity of high-dimensional data but also handles noise and outliers more effectively compared to conventional methods. In conclusion, this research provides a robust framework for data scientists and analysts aiming to implement scalable and efficient data mining solutions in diverse computational environments.