图优化的低秩双随机分解聚类

Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-20 Cooperative journals: 《计算机应用研究》

张涛 胡恩良 于景丽

Abstract： Clustering by DCD (low-rank doubly stochastic matrix decomposition) was recently proposed by Yang[16] as a method of graph clustering. DCD obtains a nonnegative low-rank doubly stochastic decomposition A=UUT(U(0) from the graph correlation matrix S by minimizing the criterion of KL (Kullback-Leibler) divergence: KL (A, S) , and clustering from U, as the class label matrix. In the method of DCD, because the S is pre-fixed, the initial value of S has a great influence on the clustering result, which leads to its lack of stability. Aiming at this problem, propose a DCD method based on graph optimization , and the optimization of graph correlation matrix S and DCD is integrated in a unified framework, which improves and extends the original DCD. The experimental results show that the graph-optimized DCD has better clustering accuracy and stability than the original DCD.

 Hits 2011  Downloads 1050  Comment