• Volume: 1
• Issue:11
• Date: 01-11-2024

Title : Comprehensive Review on Optimizing DBSCAN for Enhanced Performance in Data Mining Applications

Abstract: - Density-Based Spatial Clustering of Applications with Noise (DBSCAN) has emerged as one of the most widely used clustering algorithms in data mining due to its ability to identify clusters of arbitrary shapes and handle noise effectively. However, its performance is often challenged by parameters' sensitivity, computational complexity, and scalability with large datasets. This review paper provides a comprehensive analysis of various optimization techniques proposed to enhance DBSCAN’s performance and applicability in diverse data mining scenarios. The study examines key advancements, including parameter tuning approaches, adaptive variations, integration with parallel processing frameworks, and hybrid algorithms that combine DBSCAN with other clustering methods. Furthermore, the review highlights how these optimizations address challenges such as cluster validation, high-dimensional data handling, and real-time clustering requirements. By analyzing recent developments and comparing their efficacy, this paper offers valuable insights into the current state and future prospects of optimizing DBSCAN for data mining applications. The findings aim to guide researchers and practitioners in selecting and developing more robust and efficient clustering solutions tailored to complex data mining tasks.

Key Words: DBSCAN, density-based clustering, data mining, clustering optimization, parameter tuning, highdimensional data, parallel processing, hybrid clustering algorithms, real-time clustering

Area: Engineering

• Approved ISSN: ----
• Paper Id: IJREISTU20
• Page No: 6-10

• Author: Uvaish Akhter

---

• Co- Author: Mr. Jeetendra Singh Yadav