The Approach Based on Dividing and Conquering for Non-Negative Matrix Factorization

Non-negative matrix factorization (NMF) is useful to find basis information of non-negative data. Currently, the multiplicative updates proposed by Lee and Seung are simple and popular way to find the factorization. However, this multiplicative approach of minimizing the Euclidean distance between approximate and true values is inefficient. In this paper, firstly, the original data set were partitioned into two data subsets, and then Lee and Seung NMF approach execute on each data subset, from which the approach based on dividing and conquering for nonnegative matrix factorization was proposed. The experimental results show that the approach can effectively improve the Lee and Seung NMF approach more efficient.