Clustering algorithm based on Delaunay triangulation density metric
基于Delaunay三角剖分密度度量的聚类算法

To solve the problem that K-means clustering algorithm fails to correctly distinguish non-convex shape clusters, a clustering algorithm based on Delaunay triangulation density metric was presented. In the algorithm, Delaunay triangulation graph which has the advantages of nearest neighbor and adjacency was introduced to reflect the characteristics of data themselves and compute density. Meanwhile, chaos optimization dedicated to global optimization was applied to optimize clustering objective function for the sake of obtaining global minimum solution. Experimental results indicate that the clustering algorithm based on Delaunay triangulation density metric can find arbitrary non-convex shape clusters.