关键词: 聚类算法, 高维数据, k近邻, 密度骨架, 局部密度

Abstract: In order to find density backbone quickly and improve the accuracy of the high-dimensional data clustering results, a new algorithm based on the method of local density computation and fast recognition of high-density backbone is put forward, which is called ECLUB. Firstly, the high density backbone is identified quickly according to the mutual consistency of k neighbors and the local density relation of neighbor points. Then the unassigned low-density points are divided according to neighborhood relations to obtain the final clustering. Experiments on synthetic datasets and real datasets show the effectiveness of the proposed algorithm. The ECLUB algorithm has the advantages of high efficiency and high accuracy for high-dimensional data clustering compared with the similar algorithms.

Key words: clustering algorithm, high dimension data, k nearest neighbors, density backbone, local density