Research and Accomplishments in Applications of Non-negative Matrix Factorization

Abstract

AbstractSince its introduction in the 1990s, non-negative matrix factorization (NMF) has captured a great amount of attention due to its capability and effectiveness in processing data in such a way that few earlier methods could perform, partly due to its non-negative constraint. This paper first briefly presents the basic NMF algorithm and concerns with the algorithm itself, then demonstrates its power with three applications in three different fields, namely face recognition in Computer Vision, distance prediction in Networking and molecular pattern discovery in Genetics. The paper ends with a quick look at other applications of NMF and recent developments that researchers have made.KeywordsNon-negative matrix factorization applicationsFace recognitionNetwork distance predictionMolecular pattern discovery