Multiple Graph Adaptive Regularized Semi-Supervised Nonnegative Matrix Factorization with Sparse Constraint for Data Representation

Multiple graph and semi-supervision techniques have been successfully introduced into the nonnegative matrix factorization (NMF) model for taking full advantage of manifold structure priori information data to capture excellent low-dimensional representation. However, existing methods do not consider sparse constraint, which can enhance local learning ability improve performance in practical applications. To overcome this limitation, a novel NMF-based representation method, namely, multiple adaptive regularized semi-supervised with constraint (MSNMFSC) is developed paper obtaining discriminative increasing quality decomposition NMF. Particularly, based on standard NMF, proposed MSNMFSC method combines regularization, limited supervised together learn more parts-based Moreover, convergence analysis studied. Experiments are conducted several image datasets clustering tasks, results shown that achieves better than most related methods.