Density Peaks Clustering Based on Improved RNA Genetic Algorithm

A density peaks clustering based on improved RNA genetic algorithm (DPC-RNAGA) is proposed in this paper. To overcome the problems of Clustering by fast search and find of density peaks (referred to as DPC), DPC-RNAGA uses exponential method to calculate the local density, In addition, improved RNA-GA was used to search the optimums of local density and distance. So clustering centers can be determined easily. Numerical experiments on synthetic and real-world datasets show that, DPC-RNAGA can achieve better or comparable performance on the benchmark of clustering, adjusted rand index (ARI), compared with K-means, DPC and Max_Min SD methods.