Multi Angle Analysis of The Existing Clustering Algorithms

Data mining clustering is a broad research field. It is used to partition the data set of clusters. Different clustering methods use different similarity definition and technology. Several popular clustering algorithms are analyzed from three different perspectives: the clustering criterion, clustering algorithm and frame representation. Furthermore, some new construction algorithm, mixed or generalization of some algorithm were introduced. As a result of the analysis of several points of view, it can be covered and distinguished from most existing algorithms. It is based on self tuning algorithm and clustering benchmark.