International Journal of Computer and Information Technology (IJCIT)
Co-clustering is to partition rows and columns of a matrix simultaneously. It has been an important research field in data mining and machine learning. It is preferred over traditional homogeneous clustering techniques in many real applications. In this paper, the authors present a co-clustering algorithm based on local information and regularization. The algorithm seeks to preserve the local intrinsic geometry and measure smoothness of indicator functions with respect to the bipartite graph. The minimization of the objective function can be formulated as a generalized eigenvalue problem.