A Density-Based Clustering Algorithm for High-Dimensional Data with Feature Selection

摘要

The density-based spatial clustering of applications with noise (DBSCAN) is a kind of the density-based representative algorithms. It has been widely used in more and more fields due to its ability to detect clusters of different sizes and shapes. However, the algorithm becomes unstable when dealing with the high dimensional data. To solve the problem, an improved DBSCAN algorithm based on feature selection (FS-DBSCAN) is proposed. The performance of this algorithm is testified by a series of simulations on real world datasets. Comparisons with the DBSCAN algorithm demonstrate the superiority of the proposed algorithm.