Data Mining using Matrices and Tensors (DMMT'08)
Chris H. Q. Ding · 2008
Clustering is a very important topic in machine learning and knowledge discovery research. Many methods have been proposed, each based on different assumptions and models. In this paper we propose an improvement of the Principal Direction Divisive Partitioning algorithm. The proposed algorithm merges concepts from density estimation and projection-based methods towards a fast and efficient clustering algorithm, capable of dealing with high dimensional data. Experimental results show improved partitioning performance compared to other popular methods. Moreover, we explore the problem of automatically determining the number of clusters that is central in cluster analysis.