Research on L21 Incremental Non-Negative Matrix Factorization with Sparsity Constraints
In order to solve the phenomenon of the increasing opertion efficiency caused by increasing new data,an improved algorithm of incremental non-negative matrix factorization with sparsity constraints is proposed,and the algorithm uses the L21 norm for incremental data with the addition a sparse condition.Firstly,classical non-negative matrix factorization is performed on initial data,and then its factorization results are used to participate in the operation of the incremental data,so that the objective function has a better convergence effect in the calcula-tion of factorization and a better sparsity of the data after factorization.In the experiment,the proposed algorithm is compared with incremental non-negative matrix factorization,incremental non-negative matrix factorization with sparsity constraints and classical non-negative matrix factorization algorithms,and it is concluded that the proposed algorithm is better than the other three algorithms in terms of the sparsity and convergence speed of the data after factorization.
NETL
NSTL
万方数据
