A novel bidirectional clustering algorithm based on local density

Abstract With the widely application of cluster analysis, number clusters is gradually increasing, as difficulty in selecting judgment indicators numbers. Also, small are crucial to discovering extreme characteristics data samples, but current clustering algorithms focus mainly on analyzing large clusters. In this paper, a bidirectional algorithm based local density (BCALoD) proposed. BCALoD establishes connection between points density, can automatically determine clusters, more sensitive and reduce adjusted parameters minimum. On basis robustness noise, denoising method suitable for Different cutoff distance assigned each cluster, which results improved performance. Clustering ability verified by randomly generated datasets city light satellite images.