Hybrid clustering for large sequential data

We propose a hybrid clustering algorithm for sequential data, which combines medoid-based partitioning and agglomerative hierarchial clustering. This algorithm works efficiently by inheriting partitioning clustering strategy and operates effectively by following hierarchial clustering. The proposed algorithm is designed by taking into account the specific features of sequential data modeled in metric space. More importantly, it requires O(n?n) in total to manage an iterative pre-partitioning process and a natural neighbor inspired merging process. Experimental results demonstrate the virtue of our approach.