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.
KeywordsRNA genetic algorithm Clustering Density peaks Distance
This research is supported by Excellent Young Scholars Research Fund of Shandong Normal University, China. It is also supported by Natural Science Foundation of China (No. 61472231, No. 61640201, No. 61502283, No. 61402266). And in part by the Jinan Youth Science and Technology Star Project under Grant 20120108, and in part by the soft science research on national economy and social information of Shandong, China under Grant(2015EI013).