Name
Abstract | ||
|---|---|---|
We propose a novel nonnegative matrix factorization model that aims at finding localized, part-based, representations of nonnegative multivariate data items. Unlike the classical nonnegative matrix factorization (NMF) technique, this new model, denoted "nonsmooth nonnegative matrix factorization” (nsNMF), corresponds to the optimization of an unambiguous cost function designed to explicitly represent sparseness, in the form of nonsmoothness, which is controlled by a single parameter. In general, this method produces a set of basis and encoding vectors that are not only capable of representing the original data, but they also extract highly localized patterns, which generally lend themselves to improved interpretability. The properties of this new method are illustrated with several data sets. Comparisons to previously published methods show that the new nsNMF method has some advantages in keeping faithfulness to the data in the achieving a high degree of sparseness for both the estimated basis and the encoding vectors and in better interpretability of the factors. |
Year | DOI | Venue |
|---|---|---|
2006 | 10.1109/TPAMI.2006.60 | IEEE Trans. Pattern Anal. Mach. Intell. |
Keywords | Field | DocType |
constraint optimization,nonnegative matrix factorization,pattern analysis,sparse matrices,design optimization,encoding,matrix decomposition,artificial intelligence,data mining,face,electroencephalography,cost function,feature extraction,constrained optimization,computer simulation,indexing terms,independent component analysis,multivariate data,algorithms,datamining,vectors | Interpretability,Data set,Pattern recognition,Nonnegative matrix,Computer science,Matrix decomposition,Feature extraction,Artificial intelligence,Non-negative matrix factorization,Sparse matrix,Constrained optimization | Journal |
Volume | Issue | ISSN |
28 | 3 | 0162-8828 |
Citations | PageRank | References |
153 | 8.69 | 14 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Alberto Pascual-Montano | 1 | 590 | 32.41 |
| José María Carazo | 2 | 654 | 56.25 |
| Kieko Kochi | 3 | 182 | 11.20 |
| Dietrich Lehmann | 4 | 240 | 21.63 |
| Roberto D. Pascual-marqui | 5 | 378 | 25.45 |