| As an important research topic in the field of clustering algorithm of data mining,subspace clustering algorithm has a wide range of applications in text analysis,machine learning,computer vision,information retrieval,and other fields.However,the existing subspace clustering method are often based on hard partitioning techniques,and only allow data sample belongs to a single category,the clustering accuracy is low when there is overlap between the clusters.Aiming at this problem,a novel subspace overlapping clustering algorithm based on elastic-net(SOC-EN)is proposed.Firstly,the high-dimensional data is divided into several subspaces by using the subspace representation of elastic-net.Then,a probability model of the exponential family distribution is used to determine the overlapping part of the clusters in the subspace,and the data is assigned to the correct class clusters to get the clustering results.An alternating maximization method is used to determine the optimal solution of the objective function in the process of parameter estimation.Experiments on artificial datasets and internationally recognized UCI datasets show that the algorithm has better clustering performance by being compared to other contrast algorithms and it is suitable for large scale datasets. |
