Abstract
Clustering by fast search and find of density peaks (DP) is a method in which density peaks are used to select the number of cluster centers. The DP has two input parameters: 1) the cutoff distance and 2) cluster centers. Also in DP, different methods are used to measure the density of underlying datasets. To overcome the limitations of DP, an Adaptive-DP method is proposed. In Adaptive-DP method, heat-diffusion is used to estimate density, cutoff distance is simplified, and novel method is used to discover exact number of cluster centers, adaptively. To validate the proposed method, we tested it on synthetic and real datasets, and comparison are done with the state of the art clustering methods. The experimental results validate the robustness and effectiveness of proposed method.