Abstract:
Along with the expansion and in-depth of the application domain of cluster analysis, one kind of new cluster algorithm called Spectral Clustering algorithm has been aroused great concern by scholars, Spectral Clustering algorithm is newly developing technique in the field of machine learning in recent years. Unlike the traditional clustering algorithms, this can solve the clustering of non-convex sphere of sample spaces and has globally optimal solution. This paper introduces the principle, the induction summary to the current research situation of Spectral Clustering algorithm as well as in various application domains. Firstly, the analysis and induction of some Spectral Clustering algorithms have been made from several aspects, such as the ideas of algorithm, key technology, advantage and disadvantage. On the other hand, some typical Spectral Clustering algorithms have been selected to analyze and compare. Finally, it points out the key problems and future directions.