A Scalable Clustering Algorithm Based on Affinity Propagation and Normalized Cut

In this paper, a new scalable clustering method named “APANC” (Affinity Propagation And Normalized Cut) is proposed. During the APANC process, we firstly use the “Affinity Propagation” (AP) to preliminarily group the original data in order to reduce the data-scale, and then we further group the result of AP using “Normalized Cut” (NC) to get the final result. Through such combination, the advantages of AP in time cost and the advantages of NC in accuracy have been adopted. The experimental results show that even though the proposed method includes two clustering processes, APANC is much faster than AP; at the same time, the clustering quality of APANC is comparable to that of NC. Furthermore, the advantages of APANC in time cost could be greater when data scale increases.