On αα-divergence based nonnegative matrix factorization for clustering cancer gene expression data

SummaryObjectiveNonnegative matrix factorization (NMF) has been proven to be a powerful clustering method. Recently Cichocki and coauthors have proposed a family of new algorithms based on the αα-divergence for NMF. However, it is an open problem to choose an optimal αα.Methods and materialsIn this paper, we tested such NMF variant with different αα values on clustering cancer gene expression data for optimal αα selection experimentally with 11 datasets.Results and conclusionOur experimental results show that α=1α=1 and 2 are two special optimal cases for real applications.