CFSFDP (clustering by fast search and find of density peaks) is recentlydeveloped density-based clustering algorithm. Compared to DBSCAN, it needs lessparameters and is computationally cheap for its non-iteration. Alex. at al havedemonstrated its power by many applications. However, CFSFDP performs not wellwhen there are more than one density peak for one cluster, what we name as "nodensity peaks". In this paper, inspired by the idea of a hierarchicalclustering algorithm CHAMELEON, we propose an extension of CFSFDP,E_CFSFDP, toadapt more applications. In particular, we take use of original CFSFDP togenerating initial clusters first, then merge the sub clusters in the secondphase. We have conducted the algorithm to several data sets, of which, thereare "no density peaks". Experiment results show that our approach outperformsthe original one due to it breaks through the strict claim of data sets.
展开▼